serum albumin as a carrier of immunostimulatory
TRANSCRIPT
Serum Albumin as a Carrier of Immunostimulatory Oligodeoxynucleotides for Cancer
Therapy
Patrick C. Guley
A dissertation submitted to the faculty of the University of North Carolina at Chapel Hill in
partial fulfillment of the requirements for the degree of Doctor of Philosophy in the School of
Pharmacy.
Chapel Hill
2013
Approved by:
Moo J. Cho, Ph.D.
Michael Jay, Ph.D.
Geoff Hird, Ph.D.
Leaf Huang, Ph.D.
Michael Jarstfer, Ph.D.
ii
ABSTRACT
PATRICK C. GULEY: Serum Albumin as a Carrier of Immunostimulatory Oligonucleotides
for Cancer Therapy
(Under the direction of Moo J. Cho, Ph.D.)
Toll-like receptor 9 (TLR9) is an endosomal receptor expressed on immune cells.
The receptor recognizes microbial DNA, which contains a higher frequency of unmethylated
CpG sequences than host DNA. Oligodeoxynucleotides containing such CpG motifs (CpG)
are potent activators of TLR9, causing release of inflammatory cytokines and initiating both
the innate and adaptive immune response. CpG has been successfully used to treat solid
tumors, but its use is limited by its unfavorable pharmacokinetics. In preclinical studies,
CpG was effective only when injected peritumorally; i.e., ineffective when administered
systemically. In an attempt to overcome this limitation, CpG was derivatized with a
maleimide moiety to allow a chemical reaction with free thiols. The derivative is referred to
as CpG-mal throughout this dissertation.
Albumin, the most abundant serum protein, contains one free thiol, Cys34, which
comprises greater than 80% of the serum thiol content. Consequently, upon intravenous
injection, the CpG-mal will react with circulating albumin to form a 1:1 conjugate in a
predictable manner. Serum pharmacokinetics and biodistribution of phosphodiester CpG-
mal was investigated in tumor-bearing mice using a PET/CT imaging procedure. For this
series of studies, a novel tyrosine-containing crosslinker was synthesized to facilitate
radioiodination of the CpG with 124
I in high yields. Imaging studies revealed the reaction
iii
between the CpG-mal and albumin was fast enough to outcompete the high renal clearance of
CpG, with the reaction complete within minutes. The new CpG-albumin conjugate displayed
a similar distribution pattern and plasma half-life as albumin; half-life was increased 70-fold
and tumor accumulation was increased 30-fold over control CpG.
In vitro plasma stability studies showed that albumin conjugation lead to a 1.5-fold
reduction in the rate of enzymatic degradation of phosphodiester CpG. An in vitro
macrophage activation assay indicated that phosphodiester CpG-albumin conjugates were
weak agonists of TLR9. However, phosphorothioate CpG-albumin adducts were able to
interact with TLR9 to initiate cytokine release from J774 macrophage cells, although this
activity was reduced compared to control phosphorothioate CpG. The activation was
independent of crosslinker length, and introducing a reducible disulfide crosslinkage did not
enhance the activity. Pharmacokinetics and biodistribution of phosphorothioate CpG were
measured using [3H]-labeled CpG. The [
3H]-CpG-mal had a longer plasma half-life than
control CpG, however, the liver accumulation was significantly increased. This liver uptake
led to a less striking increase in tumor accumulation of CpG-mal than control CpG.
In vivo tumor growth inhibition studies using a CT26 colon carcinoma model showed
both the CpG-mal and control CpG were equally efficacious and caused complete tumor
regression in 6/10 mice. On the other hand, in the 4T1 model no tumor regression was
observed presumably due to lack of tumor-associated antigens from the 4T1 cells.
iv
To the pursuit of Truth
v
ACKNOWLEDGEMENTS
No man is an island; and none of this would have been possible without a
considerable amount of direct and indirect assistance by many people. Obviously, my
advisor Dr. Cho has had a huge impact on my training and development as a scientist and an
intellectual (I use the word liberally). He gave me the freedom to experiment which often
times lead to failure but he was always offering encouragement and challenging me to do
better. My only hope is that the time spent was not a one-way street.
I would like to thank the members of my committee for providing insight and
assistance with designing experiments. Additionally, I want to thank all the professors in the
School of Pharmacy for teaching me diverse knowledge in the field of pharmaceutics. The
staff and administrative assistants (Kathryn, Kim, Jubina, Ning, Rod, Amber, and Ain)
deserve praise for making the bureaucratic red tape manageable and enabling quick ordering
of supplies.
Additional acknowledgment goes to members of core facilities at UNC who provided
valuable technical assistance: Arlene Bridges for mass spectrometry assistance; Karl
Koshlap for NMR assistance; Nick Shalosky at Tissue Culture Facility for cell lines; Hong
Yuan, Kevin Guley, and Carla Johnson at Biomedical Research Imaging Center for help with
imaging experiments; Charlene Santos at Animal Study Core for help with animal
experiments.
vi
I would like to thank the other graduate students and research scientists in the School
of Pharmacy for the many discussions and friendships we’ve had. I would like to thank Feng
Liu for letting me use his cell hood and Yang Liu for helping me on short notice with some
final experiments. I especially want to thank former Cho lab members: Kevin Han, Kayla
Knilans, Roland Cheung, John An, Brad Gustafson, and June Lee. Tip of the hat to Shyam
Joolakankti for teaching me tips and tricks of chemical synthesis. I particularly want to
acknowledge my friendship with Michael Hackett, we’ve had plenty of deep scientific and
philosophical conversations in addition to the many memories outside of lab.
I would like to thank my parents for their support during the many trials and
tribulations I encountered during the past few years and most importantly for providing the
many educational opportunities that led me here.
I am indebted to my wife, Natalie, whom I met during my first year of graduate
school. She is the constant balancing force in my life and has provided unconditional
positive support. I can’t wait for the birth of our first daughter in the coming weeks.
Lastly, I want to reiterate that this work is based on other damn good scientist’s
previous discoveries; many of which are now taken for granted but without them I would
have been utterly clueless. Here goes nothing…(inhale)...
vii
TABLE OF CONTENTS
ABSTRACT ........................................................................................................................ ii
ACKNOWLEDGEMENTS .......................................................................................................v
LIST OF TABLES .....................................................................................................................x
LIST OF FIGURES ................................................................................................................. xi
LIST OF ABBREVIATIONS AND SYMBOLS .................................................................. xiii
CHAPTER I: INTRODUCTION ..........................................................................................1
1.1 Cancer and Immunotherapy ........................................................................... 1
1.2 Biological Response Modifiers ...................................................................... 4
1.3 CpG Oligodeoxynucleotides .......................................................................... 6
1.4 Serum Albumin .............................................................................................. 8
1.5 Albumin and Cancer .................................................................................... 11
1.6 Proposed Studies .......................................................................................... 12
REFERENCES .............................................................................................15
CHAPTER II: SYNTHESIS AND APPLICATION OF A
HETEROTRIFUNCTIONAL CROSSLINKER FOR 124
I-
BASED PET IMAGING...............................................................................20
2.1 Overview ...................................................................................................... 20
2.2 Introduction .................................................................................................. 20
2.3 Experimental Procedures ............................................................................. 22
2.3.1 Crosslinker Synthesis ................................................................................... 22
2.3.2 ODN Chemistry ........................................................................................... 25
2.3.3 Pharmacokinetic Experiments ..................................................................... 27
2.4 Results .......................................................................................................... 29
2.4.1 Crosslinker Synthesis ................................................................................... 29
2.4.2 ODN Preparation and Iodination ................................................................. 30
2.4.3 Image Analysis............................................................................................. 30
viii
2.5 Discussion .................................................................................................... 31
REFERENCES .............................................................................................38
CHAPTER III: PHARMACOKINETICS/BIODISTRIBUTION AND
PHARMACODYNAMICS OF MALEMIDE DERIVATIZED
OLIGODEOXYNUCLEOTIDES WITH PHOSPHODIESTER
BACKBONE IN TUMOR BEARING MICE ..............................................41
3.1 Overview ...................................................................................................... 41
3.2 Introduction .................................................................................................. 41
3.3 Experimental Procedures ............................................................................. 43
3.3.1 CpG ODN Chemistry ................................................................................... 43
3.3.2 Pharmacokinetics and Biodistribution ......................................................... 45
3.3.3 In Vitro Plasma Stability .............................................................................. 48
3.3.4 In Vitro Macrophage Activation .................................................................. 50
3.4 Results .......................................................................................................... 51
3.4.1 In Vitro Plasma Stability .............................................................................. 51
3.4.2 Plasma Pharmacokinetics............................................................................. 52
3.4.3 Biodistribution and Tumor Accumulation ................................................... 52
3.4.4 4T1 Tumor Growth Study ............................................................................ 53
3.4.5 In Vitro Cytokine Release ............................................................................ 53
3.5 Discussion .................................................................................................... 54
REFERENCES .............................................................................................63
CHAPTER IV: PHARMACOKINETICS/BIODISTRIBUTION AND
PHARMACODYNAMICS OF MALEMIDE-DERIVATIZED
OLIGODEOXYNUCLEOTIDES WITH
PHOSPHOROTHIOATE BACKBONE IN TUMOR-BEARING
MICE.............................................................................................................65
4.1 Overview ...................................................................................................... 65
4.2 Introduction .................................................................................................. 65
4.3 Experimental Procedures ............................................................................. 67
4.3.1 In Vitro Cytokine Release ............................................................................ 69
4.3.2 Pharmacokinetic and Biodistribution Study ................................................ 71
4.3.3 CT26 Tumor Growth Study ......................................................................... 72
4.3.4 4T1 Tumor Growth Study ............................................................................ 73
ix
4.4 Results .......................................................................................................... 73
4.4.1 In Vitro Macrophage Activation .................................................................. 73
4.4.2 Pharmacokinetics and Biodistribution ......................................................... 74
4.4.3 Tumor Growth Studies ................................................................................. 74
4.5 Discussion .................................................................................................... 75
REFERENCES .............................................................................................88
CHAPTER V: CONCLUSIONS & FUTURE DIRECTION ..................................................91
REFERENCES .............................................................................................98
APPENDIX A .....................................................................................................................100
REFERENCES .....................................................................................................................113
x
LIST OF TABLES
Table 3-1. Plasma and tumor exposure of the [124
I]-labeled CpG. ........................................ 58
Table 4-1. Two-compartment model parameters for [3H]-labeled CpG. ............................... 82
xi
LIST OF FIGURES
Figure 1.1. The cellular mechanisms of immune activation by CpG ODN. ......................14
Scheme 2.1. Synthetic scheme for the synthesis of the heterotrifunctional
crosslinker used throughout this study. ............................................................34
Figure 2.2. Chemical structures of the modified ODN synthesized for the study. ............35
Figure 2.3. Normalized whole body PET/CT images of biodistribution of the 124
I
radiolabeled treatments after 20 min post injection. ........................................36
Figure 2.4. Time activity curves for the blood and urine during 1 h post injection. ..........37
Figure 3.1. In vitro plasma stability of PO [3H]-CpG. .......................................................56
Figure 3.2. Blood concentration of 124
I labeled CpG. ........................................................57
Figure 3.3. Tumor time activity curve showing tumor uptake of 124
I labeled CpG. ..........59
Figure 3.4. Terminal biodistribution of 124
I-labeled CpG measured by ex vivo
gamma counting. ..............................................................................................60
Figure 3.5. Survival after surgical resection of 4T1 primary tumor. ..................................61
Figure 3.6. In vitro IL-12 and IL-6 release from J444 cells. ..............................................62
Figure 4.1. Chemical structures and synthetic scheme for the synthesis of CpG-
mal and CpG-COOH........................................................................................79
Figure 4.2. In vitro IL-12 and IL-6 release from CpG-albumin conjugates. ......................80
Figure 4.3. Blood concentration of [3H]-CpG. ...................................................................81
Figure 4.4. Tumor accumulation of PS [3H]-CpG..............................................................83
Figure 4.5. Biodistribution of [3H]-labeled CpG. ...............................................................84
Figure 4.6. Tumor growth inhibition of 4T1 tumors. .........................................................85
Figure 4.7. CT26 tumor growth curves. .............................................................................86
Figure 4.8. Individual growth curves for CT26 tumors. ....................................................87
xii
Figure A.1. 1H NMR spectrum of Mal-Tyr(tBu)-OtBu (1)...............................................101
Figure A.2. 13
C NMR of Mal-Tyr(tBu)-OtBu (1) .............................................................102
Figure A.3. 1H NMR of Mal-Tyr (2) .................................................................................103
Figure A.4. 13
C NMR of Mal-Tyr (2)................................................................................104
Figure A.5. Mass Spectrum of Mal-Tyr (2) ......................................................................105
Figure A.6. 1H NMR spectrum of Mal-Tyr-TEG-COOH (3) ...........................................106
Figure A.7. 13
C NMR spectrum of Mal-Tyr-TEG-COOH (3) ..........................................107
Figure A.8. Mass spectrum of Mal-Tyr-TEG-COOH (3) .................................................108
Figure A.9. 1H NMR of Mal-Tyr-TEG-NHS (4) ..............................................................109
Figure A.10. 13
C NMR of Mal-Tyr-TEG-NHS (4) .............................................................110
Figure A.11. Mass spectrum of Mal-Tyr-TEG-NHS (4) ....................................................111
Figure A.12. Deconvoluted mass spectrum of ODN-mal. ..................................................112
xiii
LIST OF ABBREVIATIONS AND SYMBOLS
ACN Acetonitrile
AcOH Acetic Acid
APC Antigen presenting cell
BD Biodistribution
BSA Bovine Serum albumin
CpG Cytidine-phosphate-Guanosine
CT Computer assisted tomography
DCC N,N’-Dicyclohexylcarbodiimide
DCM Dichloromethane
DIEA N,N-Diisopropylethylamine
DNP 2,4 Dinitrophenol
EDC N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide
EDTA Ethylenediaminetetraacetic acid
ELISA Enzyme Linked Immunosorbent Assay
EMCS N-(ε-Maleimidocaproyloxy)succinimide
EtOAc Ethyl acetate
EtOH Ethanol
g Gravitational constant
HPLC High Performance Liquid Chromatography
HSA Human Serum Albumin
IL-6 Interleukin 6
xiv
IL-12 Interleukin 12
iv intravenous
LPS Lipopolysaccharide
mal maleimide
MeOH methanol
MSA Mouse Serum albumin
NEM N-ethylmaleimide
NHS N-hydroxysuccinimide
NK Natural Killer cell
ODN Oligodeoxynucleotide
PAMP Pathogen-associated molecular pattern
PBS Phosphate Buffered Saline
PD Pharmacodynamics
PET Positron Emission Tomography
PK Pharmacokinetics
PO Phosphodiester
PS Phosphorothioate
sc subcutaneous
TEA Triethylamine
TEAA Triethylammonium acetate
TFA Trifluroacetic acid
THF Tetrahydrofuran
TLC Thin Layer Chromatography
xv
TLR9 Toll-like receptor 9
TMS Tetramethyl silane
Tyr tyrosine
CHAPTER I: INTRODUCTION
For every minute of 2013, it is predicted that one United States citizen will succumb
to cancer.(Siegel, Naishadham et al., 2013) The total direct and indirect economic impact of
cancer-related healthcare in the United States for 2008 totaled $200 billion and will
undoubtedly rise in the future.(Society, 2013) Society needs better treatments for cancer,
especially for younger patients. Ideally, these treatments should be as cost effective as
possible. One potential way to reduce the cost of therapy and increase the effectiveness is to
utilize existing natural defense mechanisms to our advantage.
1.1 Cancer and Immunotherapy
Current medical understanding of the pathophysiology of cancer suggests we are
constantly under assault from cancer causing agents, be they ultraviolent rays from the sun,
chemical carcinogens, radiation exposure, reactive oxygen species, or viral and bacterial
infections. Any of these agents can cause the genetic mutations that transform normal cells
into cancer cells. However, as these mutations occur, the human body is not constantly
developing clinical tumors. This implicates the existence of a controlling mechanism. The
cancer-controlling mechanism largely consists of two components: an intracellular and an
extracellular mechanism.
Cell growth is normally regulated tightly by many tumor suppressor genes and
contact inhibition; furthermore, significant aberrant mutations lead to apoptosis.(Norbury and
2
Hickson, 2001) The majority of transformed cells are thought to be repaired or eliminated by
these intracellular mechanisms. Some cells can accumulate mutations in these controlling
genes and are rendered neoplastic.
The extracellular phase is known as immunosurveillance and, more recently,
immunoediting(Dunn, Old et al., 2004); the immune system is able to differentiate and
destroy these abnormal cells before they develop into a tumor. A number of experimental
studies using genetically-engineered knockout mice with incomplete immune systems have
an increase in tumor incidence. (Shankaran, Ikeda et al., 2001; Enzler, Gillessen et al., 2003;
Street, Hayakawa et al., 2004) Additionally, it is known that humans with
immunodeficiencies have a greater chance of developing tumors.(Rabkin, Biggar et al., 1991;
Grulich, van Leeuwen et al., 2007) In some instances, abnormal cells are able to escape this
control mechanism and their unchecked growth leads to cancer pathology. The exact nature
of how some cells escape the immunoediting mechanism is not currently well understood.
One school of thought involves a selective proliferation of non-immunogenic cells lacking
specific tumor antigens. The consensus view of tumor immunobiology is that the lack of
immune response is not merely due to an absence of immune cells in the vicinity of the
tumor. In fact, most tumors contain large amounts of immune cells.(Lin and Pollard, 2004)
Rather, the issue appears to be that the cells have been anergized and no longer function as
active immune cells. They have been regulated by a sub population of T cells commonly
known as Treg and other immunosurpessive cells that tell them that the tumor is not a
threat.(Wolf, Wolf et al., 2003)
There is a growing interest in using the immune system to detect and cure cancer, but
the concept is not new. Some of the earliest clinical uses of immunotherapy were performed
3
in the late 1890s by Dr. William Coley, who recognized that patients who suffered from
bacterial infections showed a regression in tumor size. He hypothesized that bacteria was the
cause of tumors, and could be treated by injection of bacterial extracts. His extracts
contained two strains of bacteria and were injected in or around the tumor. As a result, the
patients would become febrile and develop flu-like symptoms. In cases that responded
positively, the tumors would quickly undergo liquefactive necrosis and begin to dissipate.
Repeated injections over the span of multiple months provided complete remission for his
patients who initially had a positive response. In patients that did not respond to the initial
treatment, subsequent treatments were also ineffective, and these patients were given
different therapies.
Cancer vaccines are another type of heavily investigated- and thus heavily invested-
immunotherapy. This approach appears flawed because, by their very nature, cancers are
heterogeneous and thus do not express the same antigens across the population. The second
issue with cancer vaccines is that clinicians have no way of knowing which antigens are
present without a biopsy. Additionally, tumors contain a heterogeneous population of cells.
A single cancer vaccine can target a single antigen, but will not affect the tumors cells which
do not express the specific antigen. This would require a multi-vaccine approach, which
further increases cost and feasibility issues. Finally, from a philosophical standpoint, it is
questionable whether researchers can select an appropriate set of antigens in vitro against
tumors better than the patient’s own immune system in vivo.
In spite of these conceptual flaws, from a cost-benefit perspective, immunotherapies
are attractive mainly because they have the ability to exert an effect after the treatment has
cleared the body. This should translate to reduced treatment frequency and associated costs.
4
1.2 Biological Response Modifiers
Humans have evolved in intimate contact with bacteria. Our immune systems have
thus developed methods to detect and limit the threat bacteria pose to our survival. A
consequence of this natural progression is a system to detect the conserved components of
bacteria that renders virtually any type of bacteria to be identifiable by the immune system.
These conserved components are called pathogen-associated molecular patterns, or PAMPs.
They include components from gram negative cell walls such as lipopolysaccharides (LPS),
DNA and RNA, also flagella and other membrane-associated proteins.(Kawai and Akira,
2010) Some of the receptors that are responsible for detecting these are the Toll-like receptor
(TLR) family. When these PAMPs are detected by a TLR, the immune system becomes
activated and primed for further activity.
Bacterial DNA serves as a PAMP because it expresses higher levels of unmethylated
cytidine-phosphate-guanosine (CpG) dinucleotide sequences. In higher animals such as
mammals, CpG sequences occur at a quarter of the frequency.(Bird, 1986) The CpG regions
are usually flanked by residues that potentiate the immune response,(Krieg, Wu et al., 1998)
and most importantly the cytosine is highly methylated at the 5 position, which abolishes the
immune-stimulating activity.(Krieg, Yi et al., 1995) Due to these differences in
unmethylated CpG expression, our immune system has a way to discriminate between our
own self DNA and foreign bacterial DNA. Mammalian mitochondrial DNA also contains
high amounts of unmethylated CpG DNA. Thus, when released during blunt tissue trauma, it
can activate the immune system independently of exogenous bacteria.(Zhang, Itagaki et al.,
2010)
5
Unmethylated CpG is recognized by toll-like receptor 9 (TLR9). It is an intracellular
membrane receptor expressed almost exclusively on immune cells. TLR9 is thought to exist
as a homodimer, and the majority is found in the endoplasmic reticulum.(Latz, Verma et al.,
2007) When cells are exposed to CpG, TLR9 is localized to the endosomal membranes by a
currently unknown mechanism,(Latz, Schoenemeyer et al., 2004) whereby proteolytic
cleavage coverts the TLR9 into its active form.(Park, Brinkmann et al., 2008) Upon CpG
binding, the active form undergoes a conformational change and transmembrane binding of
MyD88 occurs.(Latz, Verma et al., 2007) MyD88 signaling eventually leads to activation of
NF-κB and release of inflammatory cytokines. These cytokines are able to initiate an innate
immune response by activating macrophages and natural killer (NK) cells and also an
adaptive response by dendritic cell activation of CD8+ T cells and activation of B cells
(Figure 1.1).(Krieg, 2003)
Biological response modifiers (BRMs) are attractive drug candidates for cancer
therapy because they are able to overcome some of the issues associated with tumor
vaccines. Since BRM are immune adjuvants, they are capable of activating the immune
system and potentiating targeting of tumor antigens by immune cells. This also means that a
poly-antigen response can be initiated, decreasing the chance that tumor cells will be able to
escape by altering their antigen expression. BRM can be used to either activate or potentiate
the immune system.(Williams, Ha et al., 1999) In addition to their use in cancer
immunotherapy, they can be also be used in various clinical pathologies such as allergies or
autoimmune disorders.(Gupta and Agrawal, 2010)
6
1.3 CpG Oligodeoxynucleotides
CpG oligodeoxynucleotides (ODN) are short 6-23 nucleotide DNA molecules that
bind to TLR9 and can mimic the effects of bacterial DNA.(Krieg, Yi et al., 1995) Random
screening showed the immune response to CpG ODN was sensitive to sequences flanking the
CpG: the optimal sequence found for mice was GACGTT, whereas for larger animals,
including humans and primates, the optimal sequence was GTCGTT.(Rankin, Pontarollo et
al., 2001)
Our bodies are not accustomed to having DNA exist outside the cells and have
mechanisms to remove this potential danger. Extracellular nucleases can rapidly cleave
DNA into nucleotides; 5’- or 3’-exonucleases cleave DNA at the terminal ends while
endonucleases can cleave anywhere along the strand. The majority of nucleases present
extracellularly and in plasma are 3’-exonucleases and to a lesser extent endonucleases and
5’-exonucleases. ODN that have a natural phosphodiester (PO) backbone are rapidly
metabolized by these nucleases and have poor pharmacokinetics (PK). This instability limits
their therapeutic potential.
Several strategies have emerged to increase the resistance of ODN to
nucleases.(Crooke, 1992) The most common has been to replace one of the oxygens in the
phosphate group with a sulfur atom; this modification is called phosphorothioate (PS). The
PS backbone prevents cleavage of the ODN by nucleases and the ODN remain intact to be
therapeutically active. However, this modification also alters the physicochemical properties
of the ODN and increases the hydrophobicity. PS ODN have extensive non-specific binding
to cell surfaces and proteins; they are known to bind to serum proteins and weakly bind to
7
albumin with a kd ~ 50-300 µM.(Srinivasan, Tewary et al., 1995) This binding is not very
tight as evidenced by the rapid distribution of ODN when injected into mice.(Sands, Gorey-
Feret et al., 1994) In the physiological setting, PS ODN have affinities to other targets which
seem to be collectively greater than that of albumin.
CpG monotherapy has been used in the preclinical setting to treat solid tumors. The
general consensus is that the CpG must be injected near the site of the tumor to elicit a
response.(Heckelsmiller, Rall et al., 2002; Nierkens, den Brok et al., 2009) If it is injected
systemically, the response is not sufficient enough to have clinical effects. This is no doubt a
consequence of the poor PK of ODN. When they are injected systemically, only a small
percentage of the dose arrives at the tumor, whereas if the dose is injected near the tumor, a
higher amount of the drug is in the tumor vicinity before it diffuses out. This suggests
delivery to the tumor is a limiting step in the therapeutic potential of systemically-injected
ODN.
Our lab has previously shown that IgG antibodies can be used as a carrier for CpG to
increase its systemic effectiveness.(Palma and Cho, 2007) CpG was derivatized with a 2,4-
Dintriophenol (DNP) hapten and were injected into mice that had been previously
immunized against DNP and therefore expressed a high anti-DNP IgG titer in their
circulation. The anti-DNP IgG were able to form monomeric immune conjugates with the
DNP-CpG and increase the plasma half-life and tumor accumulation. Consequently, the
DNP-CpG was considerably more efficacious than underivatized CpG in tumor suppression.
While this approach showed promise, it has limitations. The biggest issue is the
amount of specific IgG carrier. The effectiveness of the therapy is directly proportional to
8
the amount of IgG present.(Cheung and Cho, 2010) The average concentration of IgG in
normal adults is 12 mg/ml. This corresponds to the total amount of IgG for all antigens, but
only a small fraction of the IgG will be specific for a single antigen. This could be overcome
by creating an active immune response prior to CpG-hapten therapy, but that would require
extra time and more procedures. Because of these limitations, we decided to investigate the
potential of using a different endogenous carrier for CpG ODN that would be abundant and
accessible.
1.4 Serum Albumin
The most abundant protein in our blood is serum albumin. It has a molecular weight
of 67 kDa and bears a net -19 charge at physiological pH. In humans, the plasma
concentration is approximately 40 mg/ml and the interstitial concentration is approximately
20 mg/ml. This concentration difference gives rise to oncotic pressure which balances blood
pressure and potentiates osmotic pressure. Albumin has several other functions in the body:
it acts as a carrier for fatty acids, bilirubin, and numerous other endogenous ligands.(Peters,
1996) It also provides binding sites for numerous insoluble therapeutic agents. The amount
of albumin in the average human is approximately 350 g, and its half-life in healthy
individuals is estimated to be approximately 20 days.
At normal homeostasis albumin is thought to exist at steady state, where the rate of
production is equal to the rate of catabolism. The only site of albumin synthesis that has
been discovered is the liver.(Peters, 1996) The sites of catabolism have not been exclusively
established because of the widespread distribution of albumin, but it is thought the skin and
muscle account for the majority of albumin catabolism. In order to conserve energy by
9
preventing unwarranted catabolism, a recycling receptor, FcRn, is expressed on endothelial
cells and other phagocytic cells.(Akilesh, Christianson et al., 2007) FcRn has a higher
affinity for albumin at low pH, such as that found in the endosome, and lower affinity at
neutral pH. This pH-dependent binding is related to protonation of His residues of
FcRn.(Andersen, Dee Qian et al., 2006; Andersen and Sandlie, 2009) This allows the
albumin to be recovered from the endosome and recycled back to the surface of the cell, and
is responsible for the longer half-life that albumin and IgG have compared to other serum
proteins.(Chaudhury, Mehnaz et al., 2003)
There are several other receptors involved in the transport of albumin. Albondin, or
gp60, is an endothelial surface protein that can transcytose albumin into the extravascular
space.(Schnitzer and Oh, 1994) Other scavenger receptors, gp30 and gp18, have a higher
affinity for modified or degraded albumin and lead to degradation, primarily in the liver,
rather than transcytosis.(Schnitzer, Sung et al., 1992)
The half-life and distribution of albumin scales with species weight: in mice, the
halflife is 1.2 d, rat 2.5 d, rabbits 5.7 d, dogs 8 d, and humans 20 d.(Allison, 1960) Albumin
plasma profiles, when introduced via bolus injection, exhibit biphasic behavior. They are
indicative of an initial distribution to tissues followed by a slower clearance phase. In
humans, this distribution phase takes approximately 3 days to complete and the concentration
drops to approximately 40% of the initial value.(Peters, 1996)
The endothelial cells that constitute continuous capillaries have gaps between them
which allow plasma carrying nutrients to flow out of the arterial capillaries and into the
interstitium according to Starlings equation. Interestingly, these gaps are approximately 4
10
nm, which approximates the size of albumin and as a result, normal vasculature can restrict
the transport of albumin out of the vessels via hydrodynamic sieving.(Rippe, Rosengren et
al., 2002) The restriction is not a complete barrier to albumin escape but is sufficient to
generate and maintain an albumin concentration gradient across the vessel wall. It is not
clearly established if the intercellular leakage implied by Starlings equation is, in fact, the
albondin-mediated transcytosis which has since been discovered.
Albumin has 35 Cys residues; 34 of these are engaged in disulfides leaving one free
thiol, Cys34. Cys34, albeit located within a hydrophobic cleft of the albumin molecule, is
accessible to the surrounding solvent and accounts for up to 85% of the total free thiol
content of blood.(Kratz, Warnecke et al., 2002) Approximately 25% of Cys34 exists as a
disulfide with small molecular weight thiols such as cysteine or glutathione attached.
Despite being located on a large molecule, Cys34 is quite reactive because its pKa~5-6 is
much lower than the pKa~8 for most thiols. It thus exists as thiolate anion at physiological
pH,(Kratz, Warnecke et al., 2002) much more nucleophilic than a neutral free thiol group.
Since albumin contains only one Cys34, it can be utilized as a specific handle to
reproducibly derivatize albumin to form a 1:1 conjugate. Several of the most common thiol-
specific chemistries are maleimide, haloacetyl, and pyridyl disulfide. Maleimide groups
undergo Michael addition with thiols to create a stable thioether. Haloacetyls, commonly
iodoacetyl or bromoacetyl, undergo substitution reaction to form stable thioethers. In
contrast, pyridyl disulfides undergo substitution to form a disulfide bond that can be
reversibly reduced. Of these, the maleimide is probably the most frequently used and
currently researched, partly because the reaction is fast and no side products are generated.
11
Several studies have investigated the potential of using albumin as a carrier by
modifying drugs with maleimide groups.(Elsadek and Kratz, 2012) When the maleimide-
containing drugs are injected into animals they react covalently with Cys34 within
minutes.(Kratz, Warnecke et al., 2002) Due to this fast rate, it has been possible to inject
maleimide-modified drugs and have them react with circulating albumin without appreciable
loss. While a 1:1 conjugate may seem like a poor loading capacity, it avoids the
polyvalenancy which is known provoke an immune response and generate antibodies.(Singh,
Kaur et al., 2004) This may be even more important when using an immune-activating drug.
Additionally, the more that albumin is modified, the more it is recognized as non-native and
is subject to rapid clearance.(Stehle, Sinn et al., 1997)
1.5 Albumin and Cancer
Cancer cells are constantly dividing and thus require a high amount of nutrients to
sustain their growth. The need for sustenance requires the formation of new blood vessels,
which leads to a high rate of angiogenesis.(Folkman, 1990) Histological examinations
suggest that the blood vessels surrounding tumors are different than normal vessels- having
unusually large gaps between adjacent endothelial cells.(Hashizume, Baluk et al., 2000)
Indeed, the scientific rationale of many nanoparticle drug therapies is exploiting this
difference.(Jain and Stylianopoulos, 2010) In order to clear out an area for new growth of
tumor cells and neovascularization, tumors release various enzymes to degrade the adjacent
extracellular matrix. All of these biochemical phenomena induce an increase in permeability
for macromolecules, including albumin, at the tumor vicinity.(Matsumura and Maeda, 1986;
Yuan, Dellian et al., 1995) Many cancer patients develop cachexia, a state of severe
malnourishment, in part due to the tumor consuming a large amount of serum proteins and
12
nutrients to sustain its growth.(Stehle, Sinn et al., 1997) This leads to a loss of body weight
and low concentration of serum proteins.
1.6 Proposed Studies
Collectively, those facts and findings introduced above indicate that serum albumin
could be used as a carrier of CpG ODN. The proposed sequence of events that may happen
in order to lead to a response can be described as follows: (i) upon intravenous injection, the
majority of maleimide-modified CpG ODN will covalently bind to albumin thereby limiting
their rapid clearance and distribution; (ii) the CpG-albumin conjugate will behave similarly
to albumin and will have increased circulation half-life; (iii) the CpG-albumin will
extravasate from the tumor endothelium near the tumor periphery to a higher extent due to
local increased permeability; (iv) the CpG-albumin will be taken up by highly phagocytic
cells, such as macrophages and dendritic cells, in the tumor vicinity; (v) CpG-albumin will
bind to TLR9 and cause upregulation of cytokine release, NK cell activation, and silencing of
immunosuppressive Treg cells; (vi) antigen-presenting cells (APCs) will sample tumor
antigen and become activated, (vii) APCs will travel to lymph nodes to initiate clonal
expansion of CD8+ T cells; (viii) CD8+ T cells will infiltrate the tumor and begin to destroy
the tumor cells; and (ix) any distant metastases can be identified by the lasting immune
response. This scenario appears to be wholly consistent with the scientific knowledge well
established thus far and represents the scope of this dissertation.
Based on the aforementioned timeline several specific hypothesis can be stated and
experimentally tested. First, albumin conjugation of maleimide-CpG will increase the blood
half live compared to unconjugated CpG. Second, tumor exposure of maleimide-CpG will
13
be greater than unconjugated CpG. Lastly, the therapeutic efficacy of maleimide-CpG will
be increased compared to unconjugated CpG. Experiments contained within this dissertation
have been designed to test these stated hypotheses.
14
Figure 1.1. The cellular mechanisms of immune activation by CpG ODN. Image from Krieg, A.M. (2003) Nat
Med, 9(7), 831-5.
15
REFERENCES
Akilesh, S., G. J. Christianson, D. C. Roopenian and A. S. Shaw (2007). "Neonatal FcR
expression in bone marrow-derived cells functions to protect serum IgG from
catabolism." J Immunol 179(7): 4580-4588.
Allison, A. C. (1960). "Turnovers of erythrocytes and plasma proteins in mammals." Nature
188: 37-40.
Andersen, J. T., J. Dee Qian and I. Sandlie (2006). "The conserved histidine 166 residue of
the human neonatal Fc receptor heavy chain is critical for the pH-dependent binding
to albumin." Eur J Immunol 36(11): 3044-3051.
Andersen, J. T. and I. Sandlie (2009). "The versatile MHC class I-related FcRn protects IgG
and albumin from degradation: implications for development of new diagnostics and
therapeutics." Drug Metab Pharmacokinet 24(4): 318-332.
Bird, A. P. (1986). "CpG-rich islands and the function of DNA methylation." Nature
321(6067): 209-213.
Chaudhury, C., S. Mehnaz, J. M. Robinson, W. L. Hayton, D. K. Pearl, D. C. Roopenian and
C. L. Anderson (2003). "The major histocompatibility complex-related Fc receptor
for IgG (FcRn) binds albumin and prolongs its lifespan." J Exp Med 197(3): 315-322.
Cheung, R. and M. Cho (2010). "Importance of avidity for an endogenous drug carrier: an
antibody carrier for CpG oligonucleotides." Mol Pharm 7(4): 1338-1341.
Crooke, S. T. (1992). "therapeutic applications of oligonucleotides."
Dunn, G. P., L. J. Old and R. D. Schreiber (2004). "The three Es of cancer immunoediting."
Annu Rev Immunol 22: 329-360.
Elsadek, B. and F. Kratz (2012). "Impact of albumin on drug delivery - New applications on
the horizon." J Control Release 157(1): 4-28.
Enzler, T., S. Gillessen, J. P. Manis, D. Ferguson, J. Fleming, F. W. Alt, M. Mihm and G.
Dranoff (2003). "Deficiencies of GM-CSF and interferon gamma link inflammation
and cancer." J Exp Med 197(9): 1213-1219.
Folkman, J. (1990). "What is the evidence that tumors are angiogenesis dependent?" J Natl
Cancer Inst 82(1): 4-6.
16
Grulich, A. E., M. T. van Leeuwen, M. O. Falster and C. M. Vajdic (2007). "Incidence of
cancers in people with HIV/AIDS compared with immunosuppressed transplant
recipients: a meta-analysis." The Lancet 370(9581): 59-67.
Gupta, G. K. and D. K. Agrawal (2010). "CpG oligodeoxynucleotides as TLR9 agonists:
therapeutic application in allergy and asthma." BioDrugs 24(4): 225-235.
Hashizume, H., P. Baluk, S. Morikawa, J. W. McLean, G. Thurston, S. Roberge, R. K. Jain
and D. M. McDonald (2000). "Openings between defective endothelial cells explain
tumor vessel leakiness." Am J Pathol 156(4): 1363-1380.
Heckelsmiller, K., K. Rall, S. Beck, A. Schlamp, J. Seiderer, B. Jahrsdorfer, A. Krug, S.
Rothenfusser, S. Endres and G. Hartmann (2002). "Peritumoral CpG DNA elicits a
coordinated response of CD8 T cells and innate effectors to cure established tumors
in a murine colon carcinoma model." J Immunol 169(7): 3892-3899.
Jain, R. K. and T. Stylianopoulos (2010). "Delivering nanomedicine to solid tumors." Nat
Rev Clin Oncol 7(11): 653-664.
Kawai, T. and S. Akira (2010). "The role of pattern-recognition receptors in innate immunity:
update on Toll-like receptors." Nat Immunol 11(5): 373-384.
Kratz, F., A. Warnecke, K. Scheuermann, C. Stockmar, J. Schwab, P. Lazar, P. Druckes, N.
Esser, J. Drevs, D. Rognan, C. Bissantz, C. Hinderling, G. Folkers, I. Fichtner and C.
Unger (2002). "Probing the cysteine-34 position of endogenous serum albumin with
thiol-binding doxorubicin derivatives. Improved efficacy of an acid-sensitive
doxorubicin derivative with specific albumin-binding properties compared to that of
the parent compound." J Med Chem 45(25): 5523-5533.
Krieg, A. M. (2003). "CpG motifs: the active ingredient in bacterial extracts?" Nat Med 9(7):
831-835.
Krieg, A. M., T. Wu, R. Weeratna, S. M. Efler, L. Love-Homan, L. Yang, A. K. Yi, D. Short
and H. L. Davis (1998). "Sequence motifs in adenoviral DNA block immune
activation by stimulatory CpG motifs." Proc Natl Acad Sci U S A 95(21): 12631-
12636.
Krieg, A. M., A. K. Yi, S. Matson, T. J. Waldschmidt, G. A. Bishop, R. Teasdale, G. A.
Koretzky and D. M. Klinman (1995). "CpG motifs in bacterial DNA trigger direct B-
cell activation." Nature 374(6522): 546-549.
Latz, E., A. Schoenemeyer, A. Visintin, K. A. Fitzgerald, B. G. Monks, C. F. Knetter, E.
Lien, N. J. Nilsen, T. Espevik and D. T. Golenbock (2004). "TLR9 signals after
translocating from the ER to CpG DNA in the lysosome." Nat Immunol 5(2): 190-
198.
Latz, E., A. Verma, A. Visintin, M. Gong, C. M. Sirois, D. C. Klein, B. G. Monks, C. J.
McKnight, M. S. Lamphier, W. P. Duprex, T. Espevik and D. T. Golenbock (2007).
17
"Ligand-induced conformational changes allosterically activate Toll-like receptor 9."
Nat Immunol 8(7): 772-779.
Lin, E. Y. and J. W. Pollard (2004). "Role of infiltrated leucocytes in tumour growth and
spread." Br J Cancer 90(11): 2053-2058.
Matsumura, Y. and H. Maeda (1986). "A new concept for macromolecular therapeutics in
cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the
antitumor agent smancs." Cancer Res 46(12 Pt 1): 6387-6392.
Nierkens, S., M. H. den Brok, T. Roelofsen, J. A. Wagenaars, C. G. Figdor, T. J. Ruers and
G. J. Adema (2009). "Route of administration of the TLR9 agonist CpG critically
determines the efficacy of cancer immunotherapy in mice." PLoS One 4(12): e8368.
Norbury, C. J. and I. D. Hickson (2001). "Cellular responses to DNA damage." Annu Rev
Pharmacol Toxicol 41: 367-401.
Palma, E. and M. J. Cho (2007). "Improved systemic pharmacokinetics, biodistribution, and
antitumor activity of CpG oligodeoxynucleotides complexed to endogenous
antibodies in vivo." J Control Release 120(1-2): 95-103.
Park, B., M. M. Brinkmann, E. Spooner, C. C. Lee, Y. M. Kim and H. L. Ploegh (2008).
"Proteolytic cleavage in an endolysosomal compartment is required for activation of
Toll-like receptor 9." Nat Immunol 9(12): 1407-1414.
Peters, T. (1996). All about albumin : biochemistry, genetics, and medical applications. San
Diego, Academic Press.
Rabkin, C. S., R. J. Biggar and J. W. Horm (1991). "Increasing incidence of cancers
associated with the human immunodeficiency virus epidemic." Int J Cancer 47(5):
692-696.
Rankin, R., R. Pontarollo, X. Ioannou, A. M. Krieg, R. Hecker, L. A. Babiuk, S. van Drunen
and L. V. van den Hurk (2001). "CpG motif identification for veterinary and
laboratory species demonstrates that sequence recognition is highly conserved."
Antisense Nucleic Acid Drug Dev 11(5): 333-340.
Rippe, B., B. I. Rosengren, O. Carlsson and D. Venturoli (2002). "Transendothelial transport:
the vesicle controversy." J Vasc Res 39(5): 375-390.
Sands, H., L. J. Gorey-Feret, A. J. Cocuzza, F. W. Hobbs, D. Chidester and G. L. Trainor
(1994). "Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I.
Comparison of a phosphodiester and a phosphorothioate." Mol Pharmacol 45(5): 932-
943.
Schnitzer, J. E. and P. Oh (1994). "Albondin-mediated capillary permeability to albumin.
Differential role of receptors in endothelial transcytosis and endocytosis of native and
modified albumins." J Biol Chem 269(8): 6072-6082.
18
Schnitzer, J. E., A. Sung, R. Horvat and J. Bravo (1992). "Preferential interaction of albumin-
binding proteins, gp30 and gp18, with conformationally modified albumins. Presence
in many cells and tissues with a possible role in catabolism." J Biol Chem 267(34):
24544-24553.
Shankaran, V., H. Ikeda, A. T. Bruce, J. M. White, P. E. Swanson, L. J. Old and R. D.
Schreiber (2001). "IFNgamma and lymphocytes prevent primary tumour development
and shape tumour immunogenicity." Nature 410(6832): 1107-1111.
Siegel, R., D. Naishadham and A. Jemal (2013). "Cancer statistics, 2013." CA Cancer J Clin
63(1): 11-30.
Singh, K. V., J. Kaur, G. C. Varshney, M. Raje and C. R. Suri (2004). "Synthesis and
characterization of hapten-protein conjugates for antibody production against small
molecules." Bioconjug Chem 15(1): 168-173.
Society, A. C. (2013). "Cancer Facts & Figures 2013." Atlanta: American Cancer Society.
Srinivasan, S. K., H. K. Tewary and P. L. Iversen (1995). "Characterization of binding sites,
extent of binding, and drug interactions of oligonucleotides with albumin." Antisense
Res Dev 5(2): 131-139.
Stehle, G., H. Sinn, A. Wunder, H. H. Schrenk, S. Schutt, W. Maier-Borst and D. L. Heene
(1997). "The loading rate determines tumor targeting properties of methotrexate-
albumin conjugates in rats." Anticancer Drugs 8(7): 677-685.
Stehle, G., H. Sinn, A. Wunder, H. H. Schrenk, J. C. M. Stewart, G. Hartung, W. MaierBorst
and D. L. Heene (1997). "Plasma protein (albumin) catabolism by the tumor itself -
implications for tumor metabolism and the genesis of cachexia." Critical Reviews in
Oncology/Hematology 26(2): 77-100.
Street, S. E., Y. Hayakawa, Y. Zhan, A. M. Lew, D. MacGregor, A. M. Jamieson, A.
Diefenbach, H. Yagita, D. I. Godfrey and M. J. Smyth (2004). "Innate immune
surveillance of spontaneous B cell lymphomas by natural killer cells and gammadelta
T cells." J Exp Med 199(6): 879-884.
Williams, D. L., T. Ha, C. Li, J. H. Kalbfleisch, J. J. Laffan and D. A. Ferguson (1999).
"Inhibiting early activation of tissue nuclear factor-κB and nuclear factor interleukin 6
with (1→ 3)-β-d-glucan increases long-term survival in polymicrobial sepsis."
Surgery 126(1): 54-65.
Wolf, A. M., D. Wolf, M. Steurer, G. Gastl, E. Gunsilius and B. Grubeck-Loebenstein
(2003). "Increase of regulatory T cells in the peripheral blood of cancer patients."
Clin Cancer Res 9(2): 606-612.
Yuan, F., M. Dellian, D. Fukumura, M. Leunig, D. A. Berk, V. P. Torchilin and R. K. Jain
(1995). "Vascular permeability in a human tumor xenograft: molecular size
dependence and cutoff size." Cancer Res 55(17): 3752-3756.
19
Zhang, Q., K. Itagaki and C. J. Hauser (2010). "Mitochondrial DNA is released by shock and
activates neutrophils via p38 map kinase." Shock 34(1): 55-59.
CHAPTER II: SYNTHESIS AND APPLICATION OF A
HETEROTRIFUNCTIONAL CROSSLINKER FOR 124
I-BASED PET
IMAGING
2.1 Overview
An efficient method for radioiodinating drug-carrier conjugates where the site of
iodination is contained within the crosslinker has been developed. A heterotrifunctional
crosslinker was synthesized with terminal maleimide and N-hydroxysuccinimide ester groups
for conjugation to cargo and a centrally incorporated Tyr residue to allow facile labeling with
124I. The crosslinker was applied to amino-modified oligodeoxynucleotides (ODN) in order
to measure in vivo conjugation to Cys34 of circulating serum albumin. It was found the in
situ reaction was complete within minutes and proceeded quickly enough to dramatically
alter the clearance and distribution of the ODN. This labeling strategy could be used as a
way to introduce any isotope of radioiodine to various drug-carrier combinations bearing the
requisite functional groups.
2.2 Introduction
The ability to non-invasively measure the pharmacokinetics and biodistribution
(PK/BD) of experimental therapeutics is a promising tool for the pharmaceutical scientist for
a number of reasons. Primarily, it enables longitudinal studies involving small animals
where multiple blood/organ samplings are not experimentally feasible.(Rygh, Qin et al.,
21
2011) Additionally, since each animal can provide information at multiple time points, there
is a net reduction in the total number of experimental animals needed to obtain an equivalent
amount of information. Finally, non-invasive imaging can be translated to a clinical setting.
However, current imaging systems require a radioisotope as a label for detection.
Incorporation of the desired label into the therapeutic may not be trivial, as often there is not
an accessible labeling functionality.
Considerable research is being performed using macromolecular carriers and
nanoparticles to enhance the therapeutic potential of drugs, particularly for targeting drugs to
solid tumors.(Bae and Park, 2011) In many cases this involves crosslinking drug molecules
to larger carrier molecules. In this study, a labeling moiety is incorporated into the
crosslinker itself, rendering radiolabeling of the conjugate independent of the drug or carrier.
The purpose of these experiments was to investigate the potential of using serum albumin as
carrier of ODN. Other groups have shown that modifying therapeutics with a maleimide
functional group allows them to covalently react with the Cys34 of circulating
albumin.(Elsadek and Kratz, 2012) This approach has been extensively applied to small
molecule anticancer drugs(Chung and Kratz, 2006; Kratz, 2007) and various peptides(Leger,
Thibaudeau et al., 2004; Thibaudeau, Leger et al., 2005; Xie, Yao et al., 2010). The present
work was designed to test whether this approach would also work for ODN and to
specifically address whether the in situ reaction was fast enough to prevent the ODN from
rapid distribution and excretion.(Lau, Graham et al., 2012)
Positron emission tomography (PET) was chosen as the method to measure the
PK/BD of modified ODN due to its resolution and quantification. Normally, the preferred
isotope for PET imaging is 18
F (t1/2 = 110 min; β+ = 98%) but its short radiological t1/2 is not
22
applicable for measuring PK/BD at longer biological t1/2. Iodine has several radioactive
isotopes and the 124
I isotope (t1/2 = 4.2 d; β+
= 22%) can be used for PET imaging. In spite of
some limitations(Pentlow, Graham et al., 1996), 124
I is successfully quantified.
2.3 Experimental Procedures
All chemicals, except where noted, were purchased from EMD Sciences or Sigma
Aldrich and were ACS reagent grade or higher.
2.3.1 Crosslinker Synthesis (Scheme 2.1)
1. Mal-Tyr(tBu)-OtBu
To a 25-mL round bottom flask was added 115 mg (0.54 mmol) of N-ε-
maleimidocaproic acid, 120 mg (0.62 mmol) of N-(3-Dimethylaminopropyl)-N′-
ethylcarbodiimide hydrochloride (EDC·HCl), 75 mg (0.65 mmol) of N-hydroxysuccinimide
(NHS), and 3 mL of dichloromethane (DCM). After 15 min 200 mg (0.61 mmol) of H-
Tyr(tBu)-OtBu·HCl was added in 2 mL of DCM followed by 400 µL (2.9 mmol) of
triethylamine. After 4 h at 25°C the reaction mixture was evaporated and the residue was
dissolved in 15 mL of ethyl acetate and extracted twice with 10 mL of 1N HCl and 10 mL of
saturated brine. The organic layer was dried over MgSO4 and dried in vacuo. The residue
was purified on silica gel using 1:1 ethyl acetate:petroleum ether to give 160 mg (60%) of a
slight yellow oil. 1H NMR (400 MHz, CD3OD): δ 1.22 (m, 2H, –CH2CH2CH2–), 1.30 (s,
9H, –OC(CH3)3), 1.38 (s, 9H, –OC(CH3)3), 1.5-1.6 (m, 4H, –CH2CH2CH2–), 2.18 (t, 2H, –
CH2C=O), 2.9-3.05 (m, 2H, CHCH2–), 3.44 (t, 2H, NCH2–), 4.56 (t, 1H, CHCH2–), 6.79 (s,
2H, –CH=CH–), 6.9 (d, 2H, Ar–H,m), 7.1 (d, 2H, Ar–H,o). 13C NMR (100 MHz, CD3OD):
23
δ 174.4, 171.2, 154.2, 134.9, 134.2, 133.5, 132.2, 130.2, 129.3, 124.7, 124.2, 123.3, 37.2,
35.3, 28.1, 27.2, 27.1, 26.1, 25.2.
2. Mal-Tyr
To a 25-mL round bottom flask was added 650 mg of 1 and 8 mL of 50%
trifluoroacetic acid in DCM. After 10 h at 25°C, the solvent was evaporated under a stream
of N2 and the product was recrystallized from acetone/DCM to give 425 mg (86%) of slight
yellow crystals. mp: 161-163°C. 1H NMR (400 MHz, (CD3)2CO): δ 1.25 (m, 2H, –
CH2CH2CH2–), 1.5-1.6 (m, 4H, –CH2CH2CH2–), 2.2 (t, 2H, –CH2C=O), 2.9-3.1 (m, 2H,
CHCH2–), 3.44 (t, 2H, NCH2–), 4.7 (q, 1H, CHCH2–), 6.7 (d, 2H, Ar–H,m), 6.81 (s, 2H, –
CH=CH–), 7.08 (d, 2H, Ar–H,o), 7.32 (d, 1H, CONH–). 13C NMR (100 MHz, CD3OD): δ
174.6, 173.8, 171.4, 156.1, 134.8, 134.0, 133.4, 130.4, 129.6, 127.9, 115.6, 115.2, 114.8,
114.4, 37.2, 35.3, 28.0, 25.9, 25.1. ESI-MS (neg, MeOH): m/z 373.1 [M – H]-
3. Mal-Tyr-TEG-COOH
To a 25-mL round bottom flask was added 400 mg (1.1 mmol) of 2, 250 mg (2.2
mmol) of NHS, and 5 mL of freshly distilled tetrahydrofuran (THF). The reaction mixture
was placed in an ice bath and 220 mg (1.1 mmol) of N,N’-Dicyclohexylcarbodiimide was
added in 5 mL THF. After 10 h the reaction mixture was filtered and dried in vacuo. The
residue was dissolved in 5 mL DCM, placed in an ice bath, and 350 mg (1.3 mmol) of
carboxy-PEG4-amine was added followed by 250 µL (1.4 mmol) of N,N-
Diisopropylethylamine. After 1 h the reaction mixture was diluted with 20 mL DCM and
washed twice with 20 mL of 1N HCl and 20 mL of saturated brine. The DCM layer was
dried over MgSO4 and dried in vacuo. The residue was purified on silica gel using 1% acetic
24
acid in 1:9 methanol:DCM to give 465 mg (70%) of a yellow oil. 1H NMR (400 MHz,
CDCl3): δ 1.2 (m, 2H, –CH2CH2CH2–), 1.5-1.6 (m, 4H, –CH2CH2CH2–), 2.2 (t, 2H, –
CH2C=O), 2.6 (t, 2H, –CH2CH2CO–), 2.9 (m, 2H, CHCH2–), 3.2-3.7 (m, 18H, –
CH2CH2O–), 3.79 (t, 2H, NCH2–), 4.68 (q, 1H, CHCH2–), 6.71 (s, 2H, –CH=CH–), 6.76 (d,
2H, Ar–H,m), 7.01 (d, 2H, Ar–H,o), 7.06 (d, 1H, –CONHCH–) 7.38 (t, 1H, –CONHCH2– ).
13C NMR (100 MHz, CDCl3): δ 174.8, 173.5, 172.0, 171.1, 155.8, 134.4, 134.1, 130.6,
130.2, 127.6, 115.8, 115.6, 70.5 70.1, 66.8, 37.8, 36.2, 35.4, 28.5, 26.4, 25.2. ESI-MS (neg,
MeOH): m/z 620.2 [M – H]-
4. Mal-Tyr-TEG-NHS
To a 25-mL round bottom flask was added 200 mg (0.3 mmol) of 3, 115 mg (1 mmol)
of NHS, 80 mg (0.4 mmol) of EDC·HCl and 7 mL of DCM. The reaction was carried out at
4°C for 10 h. The reaction mixture was diluted with 20 mL of DCM and washed twice with
20 mL 1N HCl, 20 mL of saturated bicarbonate, and 20 mL saturated brine. The DCM layer
was dried using MgSO4 and was concentrated in vacuo to give 150 mg (65%) of a slight
yellow oil. 1H NMR (400 MHz, CDCl3): δ 1.2 (m, 2H, –CH2CH2CH2–), 1.5-1.6 (m, 4H, –
CH2CH2CH2–), 2.2 (t, 2H, –CH2C=O), 2.8-2.9 (m, 6H, C=OCH2CH2C=O, –CH2CH2CO–
), 2.9-3.0 (m, 2H, CHCH2–), 3.2-3.7 (m, 18H, –CH2CH2O–), 3.82 (t, 2H, NCH2–), 4.55 (q,
1H, CHCH2–), 6.28 (t, 1H, –CONHCH2– ), 6.45 (d, 1H, –CONHCH–), 6.69 (s, 2H, –
CH=CH–), 6.76 (d, 2H, Ar–H,m), 7.04 (d, 2H, Ar–H,o), 7.45 (s, 1H, COH). 13C NMR (100
MHz, CDCl3): δ 172.7, 171.1, 169.4, 167.0, 155.8, 134.4, 134.1, 130.7, 128.0, 116.0, 70.8,
70.6, 70.3, 65.9, 37.8, 36.4, 32.2, 28.4, 26.4, 25.8, 25.2. ESI-MS (pos, MeOH): m/z 741.5
[M + Na]+
25
2.3.2 ODN Chemistry
The ODN used in all experiments was a 20mer purchased from either Integrated
DNA Technologies (Coralville, IA) or from Girindus America, Inc. (Cincinnati, OH) with a
phosphodiester backbone, unless specifically stated. They were supplied as the Na+ salt
form. The sequence was TCCATGACGTTCCTGACGTT and contained a commercially
available 3’-amino modification.
HPLC Conditions
All HPLC analysis and purification was performed using Shimadzu SCL-10A system
controller with two Shimadzu LC-8A pumps connected to a Rainin Dynamax UV-C detector
and a Shimadzu C-R6A Chromatopac recorder. Solvent A was 5% acetonitrile in 10 mM
triethylammonium acetate buffer; solvent B was 100% acetonitrile. For analytical work an
Agilent Zorbax 300SB-C18 4.6 x 150mm analytical column with 5 µm particle size and a
total flow rate of 1.0 mL/min was used with the following gradient: t = 0-5 min, %B = 0; t =
5-30 min, %B = 0-25; t = 30-33 min, %B = 25-100. For purification work the same gradient
protocol was used with an Agilent Zorbax 300SB-C18 9.4 x 250mm semi-preparative
column and a total flow rate of 4 mL/min. All detection was performed at λ = 260 nm.
ODN-mal
To a 3-dram glass vial containing 13 mg (2.1 µmol) of ODN-NH2 in 2 mL of 0.1M
sodium phosphate buffer, pH = 7.4 was added 20 mg (28 µmol) of 4 in 850 µL of
acetonitrile. The reaction progress was judged complete after 1.5 h at 25°C by analytical
HPLC. The acetonitrile was evaporated under a stream of N2 and the reaction mixture was
purified using semi-preparative HPLC. The product peak was manually collected and
26
evaporated in vacuo after addition of excess 3 M sodium acetate to acidify the pH to 5.2.
The ODN was then ethanol precipitated from 0.3M sodium acetate at -20°C to give 7 mg
(56%) of ODN-mal as a Na+ salt. ESI-MS (neg, H2O) 7194.4 [M], 7211.8 [M + NH4].
ODN-COOH
To a microcentrifuge tube containing 900 µg of ODN-mal was added 200 µL of 50
mM NaOH. After 4 h at 37ºC the ODN was ethanol precipitated from 0.3 M sodium acetate
to give 840 µg (93%) of ODN-COOH as a Na+ salt.
Murine Mercaptalbumin
Mouse serum albumin (MSA) Fraction V was purchased from MP Biomedicals
(Solon, OH). Reaction with Ellman’s reagent indicated that the free thiol content of this
albumin ranged from 0.2-0.3 mol of thiol per mol of MSA.(Janatova, Fuller et al., 1968)
Mercaptalbumin was generated by the addition of 3 molar equivalents of DL-Dithiothreitol
and incubation for 5 min at 25°C.(Funk, Li et al., 2010) The unquenched reaction was
directly applied to a Sephadex® G-25 size exclusion column equilibrated with phosphate
buffered saline (PBS). The unretained fractions containing MSA, as measured by UV, were
pooled and Ellman’s test indicated the thiol content was 0.9-1.0 mol of thiol per mol of
MSA. The MSA was further purified by mini-Q strong anion exchange spin columns
(Pierce, Rockford, IL) and eluted with PBS to generate mercaptalbumin in a manner similar
to methods previously reported.(Janatova, Fuller et al., 1968)
ODN-MSA
27
Radiolabeled ODN-mal was added to 8 equivalents of mercaptalbumin in PBS for 2 h
at 25°C. The reaction mixture was loaded onto Q strong anion exchange spin columns and
eluted with increasing stepwise NaCl gradient in 20 mM phosphate buffer, pH = 7.4.
Unreacted MSA was eluted with 300 mM NaCl and the conjugate was eluted with 400 mM
NaCl. The buffer was exchanged to PBS by repetitive ultracentrifugation using 30 kDa
molecular weight cut off (MWCO) filters.
2.3.3 Pharmacokinetic Experiments
Iodination of ODN and MSA
The ODN were iodinated using pre-coated Iodogen® tubes (Pierce, Rockford, IL).
Na124
I was purchased from IBA Molecular (Richmond, VA). The Na124
I needed to be
regenerated to in order obtain reliable yields. The regeneration process consisted of
calculated addition of stock solution containing 1 mg/mL of NaI and 1 mg/mL of NaIO3 in 1
mM NaOH.(Verel, Visser et al., 2004) The general procedure was as follows: The ODN to
be labeled was dissolved in 100 µL of 100 mM sodium phosphate buffer at pH 7.4. To a pre-
rinsed iodination tube, added were Na124
I and a calculated amount of regeneration stock
containing 0.9 mol equivalent of total iodine relative to ODN. After 1 min the ODN was
added to the tube and the reaction was allowed to progress for 6 min with periodic gentle
shaking. The unquenched reaction was directly applied to a Sephadex® G-25 size exclusion
column equilibrated with PBS. The fractions containing ODN, as measured by UV, were
pooled and concentrated using ultracentrifugation with 3 kDa MWCO filters. MSA was
labeled using a similar procedure.
Mice and Tumor Model
28
All animals were handled in accordance with an approved protocol by UNC
Institutional Animal Care and Use Committee. 4T1 cells were purchased from ATCC
(Manassas, VA) and grown according to ATCC recommendations. Eight female Balb/c
mice, 18-20 g, were orthotopically inoculated with 1 x 105 4T1 cells in 50 µL PBS by
subcutaneous injection into the mammary fad pad. The mice were randomly divided into
four groups and imaging experiments began 15 days after tumor inoculation.
PET Image Acquisition
One day prior and throughout the imaging experiments mice were supplied ad libitum
drinking water supplemented with 0.1% KI to block thyroid uptake of labeled 124
I.(Verel,
Visser et al., 2004) All animals were anesthetized using isoflurane and catheterized via tail
vein. For each scan, two mice were placed on a cardboard platform on the scanning bed of a
GE VISTA eXplore scanner and secured with surgical tape. A heart and breathing rate probe
was used to monitor vitals while scanning. The mice were first imaged with a CT scan and
then were dynamically imaged with PET for 1 h. The animals were injected with 0.2-0.3
mCi of 124
I labeled material corresponding to 100 µg of ODN in 100 µL of sterile 0.22 µm
filtered PBS and the catheters were flushed with a minimal volume of normal saline. The
amount of activity remaining in the catheter and syringe was measured using a calibrated
dose calorimeter (Capintec CRC®-25R, Ramsey, NJ) and subtracted from the initial amount
to quantify the amount of injected activity.
Image Processing
The raw data was rebinned according to the following scheme: 0-10 minutes, 1-min
intervals; 10-30 minutes, 2-min intervals; and 30-60 minutes, 3-minute intervals. Images
29
were reconstructed using an attenuation correction, scatter correction, and 2D OSEM
projection using the supplied manufacturer software (MMWKS Image Software, Laboratorio
de Imagen HGUGM, Spain). The images were then loaded into AMIDE for
analysis.(Loening and Gambhir, 2003) The images were aligned using fiducial markers
placed below the scanning bed. Three-dimensional regions of interest (ROI) were manually
drawn around the heart and bladder using the CT images. The amount of PET signal
contained within a ROI was calculated and converted to percent of injected dose per mL
(%ID/mL) using appropriate conversions to correct for time decay and a cylindrical phantom
of known activity.
2.4 Results
2.4.1 Crosslinker Synthesis
The crosslinker used throughout this investigation was synthesized using
carbodiimide chemistry from the amino acid Tyr according to Scheme 2.1. A maleimide
group was first conjugated to the –NH2 terminus of di-t-butyl protected Tyr using EDC.
After subsequent t-butyl deprotection, the –COOH terminus was converted to an NHS ester
using DCC. Conjugation of this intermediate to the ODN-NH2 was attempted but no reaction
was observed. The failure was attributed to low aqueous solubility and stearic hindrance.
Subsequently, a tetra(ethylene glycol) spacer was added to the crosslinker, which
successfully enabled conjugation to ODN-NH2. The final crosslinker was synthesized in
overall yield of 25% over 4 steps.
30
2.4.2 ODN Preparation and Iodination
The crosslinker 4 was conjugated to ODN-NH2 using standard
conditions.(Hermanson, 1996) Non-thiol reactive control ODN-COOH was synthesized
from ODN-mal by hydrolysis of the maleimide group under basic conditions.(El-Sagheer,
Cheong et al., 2011) The structures of the ODN are shown in Figure 2.2. Thiol reactivity
was monitored using a mercaptohexanol HPLC shift assay; ODN-mal underwent a reaction
as evidenced by an increase in peak retention time, whereas ODN-COOH showed no change
in peak retention time when treated with mercaptohexanol.
Conjugation with the crosslinker enabled the ODN to be iodinated with 124
I using
commercially available pre-coated Iodogen® tubes in yields ranging from 80-90%. In a
control experiment, ODN-NH2 was modified with a crosslinker which did not contain a Tyr
residue. These ODN were not iodinated with the same iodination protocol; yields <0.5%.
This result suggests the labeling was specific to the Tyr and not due to non-specific labeling
of the ODN bases which can occur at elevated temperatures and extended labeling
times.(Piatyszek, Jarmolowski et al., 1988)
2.4.3 Image Analysis
PET/CT whole body images, Figure 2.3, show this difference in initial distribution of
the different treatments. MSA and MSA-ODN are restricted to the vasculature volume and
show high heart signal, whereas the ODN-COOH is rapidly cleared and distributed showing
low heart signal. ODN-mal shows a combination of these two patterns.
Time-dependent radioactivity curves of blood and urine were generated by integrating
the total PET signal contained within an ROI outlining the heart and bladder (Figure 2.4).
31
The heart is classically assumed to be highly perfused with blood; therefore blood
concentration was approximated by total heart concentration.(Bading, Horling et al., 2008)
The concentrations were dose-normalized and expressed as %ID/mL.
ODN typically have a PK profile that can be characterized by rapid elimination from
the plasma(Sands, Gorey-Feret et al., 1994), which is similar to the profile of ODN-COOH in
Figure 2.4. On the other hand, ODN-mal shows a different plasma profile. Initially there is
a steep drop, similar to ODN-COOH, which then transitions to a much slower plasma
clearance. The transition appears to be complete after approximately 8 min. To confirm
Cys34 of albumin is the major reaction product, ODN-mal was preconjugated to MSA ex vivo
prior to injection. This curve did not display a rapid plasma clearance indicating there was
no significant free ODN contamination. However, there was a delayed disappearance from
the plasma and into the bladder. MSA was used as a control in order to determine if there
was a difference between the ODN-MSA conjugates and native albumin. Both the MSA and
ODN-MSA appear to be restricted to vascular space and do not undergo a rapid distribution.
2.5 Discussion
The crosslinker synthesized for this study was specifically designed to accommodate
the following three functionalities: (i) a maleimide group for conjugation to -SH groups; (ii)
an activated ester for conjugation to -NH2 groups; and (iii) a phenol group for 124
I PET label
incorporation. Therefore, the tertiary nature of the Tyr was utilized as the scaffold for the
crosslinker. The crosslinker was conjugated to ODN-NH2 and enabled specific radiolabeling
of the Tyr. An attempt was made to iodinate ODN-mal with a phosphorothioate backbone;
however, yields were substantially lower. The phosphorothioate groups appear to interfere
32
with the oxidation.(Xie, Liang et al., 2012) This observation shows the crosslinker requires
compatibility with the drug or carrier and cannot be used indiscriminately. In this case, no
attempt was made to iodinate the crosslinker prior to conjugation with the phosphorothioate
ODN-NH2.
The cutoff for kidney filtration is approximately 40 kDa and continuous capillary
endothelial gaps are approximately 4 nm (Rippe, Rosengren et al., 2002), therefore, there is
no a priori reason to suspect any difference in renal clearance and distribution between
ODN-COOH and ODN-mal. Immediately upon injection and prior to any thiol reaction,
these two ODN should have similar disposition. However, the ODN-mal is able to undergo
thiol addition which has the potential to alter disposition. The ODN-mal blood curve should
follow the ODN-COOH curve at early timepoints and begin to transition to a new curve as
the thiol addition reaction occurs. The majority of free thiol content in the blood is due to
Cys34 of albumin(Kratz, Warnecke et al., 2002), therefore, this new curve should behave
similar to albumin. The expected biphasic nature of the ODN-mal blood curve is clearly
observed in Figure 2.4.
It appears that albumin is indeed the major reaction product based on the similar
dispositions of preconjugated ODN-MSA to that of ODN-mal. Interestingly, there was a
delayed clearance of ODN-MSA and we speculate this is may be due to an impurity in the
formulation, most likely dimer/multimer aggregates being recognized by scavenger receptors
and degraded.(Schnitzer and Bravo, 1993) It does not appear that monomeric ODN-MSA
adduct is overtly recognized by scavenger receptors because no delayed degradation was
observed in the ODN-mal curve and the plasma t1/2 is similar to MSA.
33
The blood and urine were not directly analyzed for stability of the label or integrity of
the conjugate. In preliminary experiments where the mice were not given KI-supplemented
drinking water to block thyroid uptake, the amount of thyroid uptake was greater for labeled
MSA compared to the ODN-mal (data not shown). Additionally, labeled MSA was excreted
predominantly through the urine, whereas all other treatments containing the radiolabeled
crosslinker were excreted in the both urine and feces. Taken together these suggest the
crosslinker iodination has greater in vivo dehalogenation stability than labeled protein
tyrosines, which is in agreement with other reports of prosthetic iodinations.(Vaidyanathan
and Zalutsky, 1990)
In summary, we were able to develop a strategy of labeling drug-carrier conjugates by
incorporating a Tyr residue into the crosslinker to allow direct iodination. Placing the
radiolabel within the crosslinker should not interfere with the drug or the carrier. This should
leave their biological functions intact. Maleimide-modified ODN were able to covalently
react with Cys34 of circulating albumin and the reaction was complete within minutes which
was fast enough to outcompete the rapid plasma clearance of the ODN. The conjugation to
albumin has a dramatic effect on the PK/BD of ODN. Regardless of the radioisotope of
iodine employed, the crosslinker is a promising tool due to its modular nature and could be
used to study a variety of different drug-carrier combinations in numerous experimental
settings.
34
Scheme 2.1. Synthetic scheme for the synthesis
of the heterotrifunctional crosslinker used
throughout this study.
35
Figure 2.2. Chemical structures of the modified ODN
synthesized for the study. The site of 124
I labeling is ortho
to the tyrosine phenolic hydroxyl, both mono- and di-
iodinated ODN were observed.
36
A B
C D
Figure 2.3. Normalized whole body PET/CT images of
biodistribution of the 124
I radiolabeled treatments after 20 min post
injection. Each panel shows a coronal slice through the heart of two
mice that were simultaneously imaged. Panel A: ODN-COOH, B:
MSA, C: ODN-mal, D: ODN-MSA. All images were normalized by
scaling the maximum color threshold with average injected dose.
37
Figure 2.4. Time activity curves for the blood and urine during 1 h post injection. Rapid clearance from the
blood and into the urine is observed for the control ODN-COOH. Reduced blood clearance and urine excretion
is observed for ODN-mal, ODN-MSA, and MSA. Each point represents the average of 2 mice imaged
simultaneously.
38
REFERENCES
Bading, J. R., M. Horling, L. E. Williams, D. Colcher, A. Raubitschek and S. E. Strand
(2008). "Quantitative serial imaging of an 124I anti-CEA monoclonal antibody in
tumor-bearing mice." Cancer Biother Radiopharm 23(4): 399-409.
Bae, Y. H. and K. Park (2011). "Targeted drug delivery to tumors: myths, reality and
possibility." J Control Release 153(3): 198-205.
Chung, D. E. and F. Kratz (2006). "Development of a novel albumin-binding prodrug that is
cleaved by urokinase-type-plasminogen activator (uPA)." Bioorg Med Chem Lett
16(19): 5157-5163.
El-Sagheer, A. H., V. V. Cheong and T. Brown (2011). "Rapid chemical ligation of
oligonucleotides by the Diels-Alder reaction." Org Biomol Chem 9(1): 232-235.
Elsadek, B. and F. Kratz (2012). "Impact of albumin on drug delivery - New applications on
the horizon." J Control Release 157(1): 4-28.
Funk, W. E., H. Li, A. T. Iavarone, E. R. Williams, J. Riby and S. M. Rappaport (2010).
"Enrichment of cysteinyl adducts of human serum albumin." Anal Biochem 400(1):
61-68.
Hermanson, G. T. (1996). Bioconjugate techniques. San Diego, Academic Press.
Janatova, J., J. K. Fuller and M. J. Hunter (1968). "The heterogeneity of bovine albumin with
respect to sulfhydryl and dimer content." J Biol Chem 243(13): 3612-3622.
Kratz, F. (2007). "DOXO-EMCH (INNO-206): the first albumin-binding prodrug of
doxorubicin to enter clinical trials." Expert Opin Investig Drugs 16(6): 855-866.
Kratz, F., A. Warnecke, K. Scheuermann, C. Stockmar, J. Schwab, P. Lazar, P. Druckes, N.
Esser, J. Drevs, D. Rognan, C. Bissantz, C. Hinderling, G. Folkers, I. Fichtner and C.
Unger (2002). "Probing the cysteine-34 position of endogenous serum albumin with
thiol-binding doxorubicin derivatives. Improved efficacy of an acid-sensitive
doxorubicin derivative with specific albumin-binding properties compared to that of
the parent compound." J Med Chem 45(25): 5523-5533.
Lau, S., B. Graham, N. Cao, B. J. Boyd, C. W. Pouton and P. J. White (2012). "Enhanced
extravasation, stability and in vivo cardiac gene silencing via in situ siRNA-albumin
conjugation." Mol Pharm 9(1): 71-80.
39
Leger, R., K. Thibaudeau, M. Robitaille, O. Quraishi, P. van Wyk, N. Bousquet-Gagnon, J.
Carette, J. P. Castaigne and D. P. Bridon (2004). "Identification of CJC-1131-albumin
bioconjugate as a stable and bioactive GLP-1(7-36) analog." Bioorg Med Chem Lett
14(17): 4395-4398.
Loening, A. M. and S. S. Gambhir (2003). "AMIDE: a free software tool for multimodality
medical image analysis." Mol Imaging 2(3): 131-137.
Pentlow, K. S., M. C. Graham, R. M. Lambrecht, F. Daghighian, S. L. Bacharach, B.
Bendriem, R. D. Finn, K. Jordan, H. Kalaigian, J. S. Karp, W. R. Robeson and S. M.
Larson (1996). "Quantitative imaging of iodine-124 with PET." J Nucl Med 37(9):
1557-1562.
Piatyszek, M. A., A. Jarmolowski and J. Augustyniak (1988). "Iodo-Gen-mediated
radioiodination of nucleic acids." Anal Biochem 172(2): 356-359.
Rippe, B., B. I. Rosengren, O. Carlsson and D. Venturoli (2002). "Transendothelial transport:
the vesicle controversy." J Vasc Res 39(5): 375-390.
Rygh, C. B., S. Qin, J. W. Seo, L. M. Mahakian, H. Zhang, R. Adamson, J. Q. Chen, A. D.
Borowsky, R. D. Cardiff, R. K. Reed, F. R. Curry and K. W. Ferrara (2011).
"Longitudinal investigation of permeability and distribution of macromolecules in
mouse malignant transformation using PET." Clin Cancer Res 17(3): 550-559.
Sands, H., L. J. Gorey-Feret, A. J. Cocuzza, F. W. Hobbs, D. Chidester and G. L. Trainor
(1994). "Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I.
Comparison of a phosphodiester and a phosphorothioate." Mol Pharmacol 45(5): 932-
943.
Schnitzer, J. E. and J. Bravo (1993). "High affinity binding, endocytosis, and degradation of
conformationally modified albumins. Potential role of gp30 and gp18 as novel
scavenger receptors." J Biol Chem 268(10): 7562-7570.
Thibaudeau, K., R. Leger, X. Huang, M. Robitaille, O. Quraishi, C. Soucy, N. Bousquet-
Gagnon, P. van Wyk, V. Paradis, J. P. Castaigne and D. Bridon (2005). "Synthesis
and evaluation of insulin-human serum albumin conjugates." Bioconjug Chem 16(4):
1000-1008.
Vaidyanathan, G. and M. R. Zalutsky (1990). "Protein radiohalogenation: observations on
the design of N-succinimidyl ester acylation agents." Bioconjug Chem 1(4): 269-273.
Verel, I., G. W. Visser, M. J. Vosjan, R. Finn, R. Boellaard and G. A. van Dongen (2004).
"High-quality 124I-labelled monoclonal antibodies for use as PET scouting agents
prior to 131I-radioimmunotherapy." Eur J Nucl Med Mol Imaging 31(12): 1645-
1652.
40
Xie, D., C. Yao, L. Wang, W. Min, J. Xu, J. Xiao, M. Huang, B. Chen, B. Liu, X. Li and H.
Jiang (2010). "An albumin-conjugated peptide exhibits potent anti-HIV activity and
long in vivo half-life." Antimicrob Agents Chemother 54(1): 191-196.
Xie, X., J. Liang, T. Pu, F. Xu, F. Yao, Y. Yang, Y. L. Zhao, D. You, X. Zhou, Z. Deng and
Z. Wang (2012). "Phosphorothioate DNA as an antioxidant in bacteria." Nucleic
Acids Res 40(18): 9115-9124.
CHAPTER III: PHARMACOKINETICS/BIODISTRIBUTION AND
PHARMACODYNAMICS OF MALEMIDE DERIVATIZED
OLIGODEOXYNUCLEOTIDES WITH PHOSPHODIESTER BACKBONE IN
TUMOR BEARING MICE
3.1 Overview
A 20-mer CpG oligodeoxynucleotide (ODN) with a phosphodiester backbone was
derivatized with maleimide (CpG-mal) at 3’-end to promote their reaction with Cys34 of
serum albumin. In vitro plasma stability indicated albumin conjugation could partially
protect the CpG from nuclease degradation. Pharmacokinetics (PK) and biodistribution (BD)
were measured by PET/CT imaging of [124
I]-labeled CpG-mal. Plasma and tumor exposure
of CpG-mal was increased 70- and 30-fold, respectively, compared to control CpG. In vivo
efficacy was measured in an orthotopic 4T1 murine breast carcinoma model. No difference
was observed in tumor growth, post-resection survival, or number of lung metastasis for any
treatments compared to negative control. In vitro macrophage activation was assessed by IL-
6 and IL-12 production in J774 cells. The lack of antitumor response is explained by the
weak agonist properties of CpG with phosphodiester backbone.
3.2 Introduction
Cancer immunotherapy is a promising field of cancer research that has yet to live up
to its potential. Systemic CpG monotherapy as a cancer treatment is presumably limited by
42
the low uptake by dendritic cells and macrophages in the tumor vicinity due to the poor
plasma PK and tumor accumulation of CpG. A method of increasing the plasma halflife of
systemically-administered CpG should allow more opportunity for the CpG to interact with
the tumor tissue and increase the tumor exposure, thereby improving efficacy.(Palma and
Cho, 2007) Using serum albumin as a carrier of anticancer drugs is an attractive way to
increase the plasma half-life and tumor exposure.(Elsadek and Kratz, 2012) Derivatizing
therapeutics with maleimide groups to allow them to covalently react with Cys34 of serum
albumin.(Kratz, Warnecke et al., 2002) Maleimide-modified ODN were able to covalently
bind with circulating albumin within minutes which dramatically reduced the initial
distribution of the ODN compared to control ODN which was otherwise rapidly cleared from
the blood (Figure 2.4). Importantly, the study showed that albumin could be accessed by
CpG-mal quickly enough to outcompete the rapid clearance of free CpG-mal so that ex vivo
albumin conjugation was unnecessary.
The purpose of this study is to determine the PK and BD of CpG-mal and to test the
efficacy in tumor-bearing mice. A rationale for exploring CpG immunotherapies lies in the
premise that they can lead to both an innate and an adaptive immune response that causes
primary tumor regression, prevent tumor reoccurrence, and treat distant
metastases.(Kawarada, Ganss et al., 2001; Kunikata, Sano et al., 2004) Advances in surgery
have led to a good prognosis for non-metastatic primary tumors; however, prognosis is
considerably worse after the presence of metastases. Therefore treatment of metastasis is of
paramount clinical importance.
The interaction between tissue stroma and cancer cells has an effect on the
development of solid tumors.(Tlsty and Coussens, 2006; Ma, Dahiya et al., 2009) The use of
43
xenograph models may not sufficiently replicate this delicate balance. For these reasons we
have chosen to use a most aggressive orthotopic breast cancer model that is capable of
metastasis in order to test the efficacy of the CpG therapy.
3.3 Experimental Procedures
All chemicals, except where noted, were purchased from EMD Sciences or Sigma
Aldrich and were ACS reagent grade or higher.
3.3.1 CpG ODN Chemistry
Unless specifically stated, the CpG used in all experiments were purchased from
either Integrated DNA Technologies (Coralville, IA) or from Girindus America, Inc.
(Cincinnati, OH) with a phosphodiester backbone. They are supplied as the Na+ salt form.
The CpG-NH2 sequence was CpG1826: TCCATGACGTTCCTGACGTT and contained a
commercially available 3’-amino modification; non-stimulatory GpC-NH2 sequence
CpG1982: TCCAGGACTTCTCTCAGGTT was also purchased with a commercially
available 3’-amino modification. Unmodified CpG1826 (CpG) was also purchased.
HPLC Conditions
All HPLC analysis and purification was performed using Shimadzu SCL-10A system
controller with two Shimadzu LC-8A pumps connected to a Rainin Dynamax UV-C detector
and a Shimadzu C-R6A Chromatopac recorder. Solvent A was 5% acetonitrile in 10 mM
triethylammonium acetate buffer; solvent B was 100% acetonitrile. For analytical work an
Agilent Zorbax 300SB-C18 4.6 x 150 mm analytical column with 5 µm particle size was
used with a total flow rate of 1 mL/min using the following gradient: t = 0-5 min, %B = 0; t =
44
5-30 min, %B = 0-25; and t = 30-33 min, %B = 25-100. For purification work the same
gradient protocol was used but with an Agilent Zorbax 300SB-C18 9.4 x 250 mm semi-
preparative column and a total flow rate of 4 mL/min. All detection was performed at λ =
260 nm.
CpG-mal
To a microcentrifuge tube containing CpG-NH2 in 100 mM sodium phosphate buffer
at pH = 7.4 was added 20 equivalents of N-(ε-Maeimidocaproyloxy)succinimide ester
(EMCS) (Pierce, Rockford, IL) in acetonitrile. After 90 min at 25°C the reaction was judged
complete by analytical HPLC. The mixture was concentrated under a stream of N2 and
purified using semi-preparative HPLC. The desired peak was manually collected and
concentrated in vacuo after acidification to pH = 5.2 by the addition of excess 3M sodium
acetate buffer. The CpG-mal was ethanol precipitated from 0.3 M sodium acetate to give the
Na+ salt; yields: 55-70%. ESI-MS (neg, H2O) 6461.2 [M]
CpG-COOH
To a microcentrifuge tube containing CpG-mal was added 50 mM NaOH. After 2h at
37°C the reaction was judged complete by analytical HPLC. The CpG-COOH was ethanol
precipitated from 0.3 M sodium acetate at -20°C to give the Na+ salt; yield: 97%.
CpG-MSA Conjugation
Mouse serum albumin (MSA) Fraction V was purchased from MP Biomedicals
(Solon, OH). Conjugation of CpG-mal to MSA was performed according to a previously
described procedure in Section 2.3.2. Briefly, MSA was reacted with 3 equivalents of DTT
45
for 5 min at 25°C to generate mercaptalbumin and separated on a Sephadex® G-25 size
exclusion column equilibrated with phosphate buffered saline (PBS). The void fractions
containing MSA were pooled and loaded onto a mini-Q strong anion exchange column
(Pierce, Rockford, IL) and eluted with PBS. The MSA was added in 8-fold excess to CpG-
mal in PBS. After 2 h at 25°C the reaction mixture was loaded onto a mini-Q strong anion
exchange column and eluted with increasing stepwise NaCl gradient in 20 mM sodium
phosphate buffer pH = 7.4. Unreacted MSA was eluted with 300 mM NaCl, CpG-albumin
conjugate was eluted with 400 mM NaCl, and unreacted CpG-mal was eluted with 500 mM
NaCl. When phosphorothioate CpG-mal was used a stronger gradient was needed: 500 mM
to elute unreacted MSA, 1 M NaCl to elute CpG-MSA, and 2 M NaCl to elute unreacted
CpG-mal. Buffer exchange to PBS was performed using 30 kDa molecular weight cut off
(MWCO) ultracentrifugation (Millipore, Billerica, MA); yield: 50-70%.
3.3.2 Pharmacokinetics and Biodistribution
[124
I] Radiolabeling
Na124
I was purchased from IBA Molecular (Richmond, VA) two days prior to
iodination. CpG-mal and CpG-COOH were radiolabeled with 124
I according to a previous
procedure described in Section 2.3.3. Briefly, CpG-NH2 was conjugated to a maleimide
crosslinker that contains a tyrosine residue and was radiolabelled with 124
I using Iodogen®
precoated tubes (Pierce, Rockford, IL). The iodination protocol was as follows: CpG to be
labeled was dissolved in 100 µL of 100 mM sodium phosphate buffer at pH 7.4. To a pre-
rinsed iodination tube, added were Na124
I and a calculated amount of 1 mg/mL of NaI and 1
mg/mL of NaIO3 in 1 mM NaOH containing 0.9 mol equivalent of total iodine relative to
46
CpG. After 1 min the CpG was added to the tube and the reaction was allowed to progress
for 6 min at 25°C with periodic gentle shaking. The unquenched reaction was directly
applied to a Sephadex® G-25 size exclusion column equilibrated with PBS. The fractions
containing CpG ODN, as measured by UV, were pooled and concentrated using
ultracentrifugation with 3 kDa MWCO filters. MSA was labeled using a similar procedure.
Ex vivo conjugated CpG-MSA was prepared by using 124
I labeled CpG-mal in the same
manner as previously described.
PET Image Acquisition
One day prior and throughout the imaging experiments mice were supplied ad libitum
drinking water supplemented with 0.1% KI to block thyroid uptake of labeled 124
I.(Verel,
Visser et al., 2004) All animals were anesthetized using isoflurane and catheterized via tail
vein. For each scan, two mice were placed on a cardboard platform on the scanning bed of a
GE VISTA eXplore scanner and secured with surgical tape. A heart and breathing rate probe
was used to monitor vitals while scanning. The mice were first imaged with a CT scan and
then were dynamically imaged with PET for 1 h and statically imaged at 4, 8, 16, 24, 36 hr
for 5 minutes. The animals were injected with 0.2-0.3 mCi of 124
I labeled material
corresponding to 100 µg of ODN in 100 µL of sterile 0.22 µm filtered PBS and the catheters
were flushed with a minimal volume of normal saline. The amount of activity remaining in
the catheter and syringe was measured using a calibrated dose calorimeter (Capintec CRC®
-
25R, Ramsey, NJ) and subtracted from the initial amount to quantify the amount of injected
activity.
47
Immediately after the final image acquisition the mice were sacrificed by
exsanguination by cardiac puncture under isoflurane anesthesia. In order to remove any
residual organ blood, each mouse was perfused for 4 min with fetal bovine serum containing
100 IU/mL of heparin sodium using a peristaltic pump (Rainin Rabbit Plus, Woburn MA) at
a flow rate of 1.5 mL/min by inserting and clamping a blunt needle into the left ventricle and
cutting the vena cava. The heart, lungs, liver, kidneys, spleen, and tumor were harvested,
rinsed in PBS, blotted dry and inserted into preweighed microfuge vials. The organs were
stored at -20°C until the radioactivity was measured using a well gamma counter
(PerkinElmer 2470 WIZARD2, Waltham, MA).
Image Processing
The raw data for the dynamic scan was rebinned according to the following scheme:
0-10 minutes, 1-min intervals; 10-30 minutes, 2-min intervals; and 30-60 minutes, 3-minute
intervals. Images were reconstructed using an attenuation correction, scatter correction, and
2D OSEM projection using the supplied manufacturer software (MMWKS Image Software,
Laboratorio de Imagen HGUGM, Spain). The images were then loaded into AMIDE for
analysis.(Loening and Gambhir, 2003) The images were aligned using fiducial markers
placed below the scanning bed. Three dimensional regions of interest (ROI) were manually
drawn on the heart, tumor, lungs, liver, kidneys, and intestines using the CT images. The
amount of PET signal contained within a ROI was calculated and converted to percent of
injected dose per mL (%ID/mL) using appropriate conversions to correct for time decay and
a cylindrical phantom of known activity. Non-compartmental and PK analysis were
performed using MATLAB 2012b (The MathWorks, Inc., Natick, MA).
48
3.3.3 In Vitro Plasma Stability
[3H] Radiolabeling
CpG-NH2 was radiolabelled using the procedure of Graham et al.(Graham, Freier et
al., 1993) Briefly, 2 mg of CpG-NH2 was lyophilized from 200 µL of 0.1 M sodium
phosphate buffer, pH = 7.8, containing 0.2 mM EDTA in microfuge tube. To the dry powder
was added 200 µL of T2O (5 Ci/g, Moravek Biochemicals; Brea, CA) and 8.3 µL of β-
mercaptoethanol. After 6h at 90℃ the reaction mixture was subjected to repeated
ultracentrifugation using 5kDa MWCO filters to remove the bulk of excess T2O.
Exchangeable tritium was removed by several rounds of suspension in 1 mL H2O, incubation
at 25°C for 1 h, and lyophilization. The [3H]-CpG-NH2 (1.5 mg; 75% yield) was stored as a
dry power at -20℃ until use. Analytical HPLC indicated no degradation had occurred during
the procedure and the [3H]-CpG-NH2 was processed with EMCS as above.
Plasma Incubation
Whole blood was collected from female Balb/c mice via cardiac puncture into
heparinized tubes and was spun at 2,000 x g for 10 min at 25°C. The supernatant was
collected and filtered through 0.22 µm PVDF filters and stored at 4°C until use. To a
microfuge tube containing 200 µL of plasma at 37°C was added 25 µg of [3H]-CpG ODN in
25 µL PBS. At specified timepoints, a 25 µL aliquot was removed for size exclusion
analysis. For the [3H]-CpG-COOH sample an additional 5 µL aliquot was removed for gel
electrophoresis and was added to 25 µL of urea gel loading buffer and stored at 4°C until use.
A control experiment where 25 µg of CpG-mal was incubated in 225 µL of PBS at 37°C was
also performed.
49
A 1 x 30 cm column was packed with Sephadex® G-50 and equilibrated with PBS
containing 5 mM EDTA. The plasma aliquot was diluted with 25 µL of the column running
buffer prior to column loading. Fractions were collected using a Bio-Rad model 2110
fraction collector set to 30 drops per fraction. From each fraction 550 µL was removed and
transferred to a 20 ml scintillation vial followed by the addition of 3 mL of Ultima Gold XR
scintillation cocktail fluid. The radioactivity in each vial was measured using a Packard Tri-
Carb 2900TR liquid scintillation analyzer and reported in DPM.
The amount of radioactivity in each fraction was plotted versus fraction number.
Since all fractions were of equal size, the area under the curve (AUC) was calculated by the
summing the radioactivity over an empirically determined range of fractions corresponding
to the column void and retention volumes. The amount of degradation was calculated
according to the following equation:
[
] [ ]
[ ]
The fraction degraded was plotted versus time and the data was fit using a least-squares
linear regression algorithm.
Gel electrophoresis
Gel electrophoresis was performed using 15% Mini-PROTEAN® TBE-Urea Precast Gels
(Bio-Rad, Hercules, CA) and run at 200V for 40 min. The gels were stained with SYBR®
Gold (Molecular Probes, Eugene, OR) and imaged using AlphaImager (Alpha Innotech, San
Leandro, CA) equipped with a SYBR photographic filter.
4T1 Tumor Growth Study
50
Fifty female Balb/c mice, age 6-8 weeks, were orthotopically inoculated with 1 x 105
syngeneic 4T1 cells in 50 µL PBS into the mammary fat pad. Once the tumors were
established reached a diameter of 5 mm, as measured by calipers, the mice were randomly
divided into groups of 8 and experimental treatments were initiated. A total of five tail vein
injections were given every other day consisting of 100 µg equivalents of CpG in 100 µL
PBS. All treatments were sterile filtered using 0.22 µm PVDF membrane filters prior to
injection. The experiment treatment arms were as follows: group A received 100 µg of CpG-
mal, group B received 100 µg of GpC-mal, group C received 100 µg CpG-COOH, group D
received 100 µl of PBS, group E received 100 µg of CpG. The tumors were allowed to grow
to a maximum diameter of 12 mm at which point they were surgically resected and flash-
frozen in liquid N2. The mice were allowed to continue to survive until either of the humane
endpoints of 20% loss of total body weight loss or body condition score ≤ 2 were reached.
The mice were sacrificed by cervical dislocation after CO2 asphyxiation and the lungs and
livers were harvested in formalin. Tumor volumes were calculated as V = ½ab2, where a is
longest diameter and b is the length of the perpendicular diameter in mm. Metastatic nodules
in the lungs were counted by gross visual inspection of the organ.
3.3.4 In Vitro Macrophage Activation
J774 cells were obtained from the UNC Tissue Culture Facility, with provenance
from ATTC (Manassas, VA), and were cultured with Dubelco modified essential media
supplemented with 10% bovine calf serum in 5% CO2 atmosphere at 37°C. The cells were
grown in T150 flasks and harvested by scraping when 80% confluence was reached. All
experiments were performed with cells with a passage number less than twenty. For
experiments, the entire contents of a T-150 flask would be transferred to a 24 well plate (ca.
51
7 x 105 cells/well by hemocytometer count) in 1 mL/well of media. After 4 h, the media was
removed, and replaced with 900 µL of fresh media and 100 µL of PBS containing the desired
treatment. All treatments were sterile filtered using 0.22 µm PVDF membrane filters prior to
dilution and addition to the cells. After 20 h, the media was removed and spun at 1000 x g
for 5 min at 25°C. The supernatant was carefully removed and assayed within 1 h or frozen
on dry ice and stored at -20°C until ELISA analysis.
IL-6 and IL-12 (p40) ELISA kits and appropriate buffer sets were purchased from BD
Biosciences (San Jose, CA). The assays were performed in accordance to manufacturer’s
instructions. When dilutions were necessary the cell culture supernatant was diluted with
assay diluent. The absorbance was read at 450 nm and was corrected with 570 nm
subtraction using a Hidex Plate Chameleon™V (Turku, Finland) plate reader. The
concentration of cytokine was then back calculated using the appropriate standard curve and
dilution factors. The results were assayed in duplicate and are expressed as the average.
3.4 Results
3.4.1 In Vitro Plasma Stability
The stability of radiolabeled [3H]-CpG-COOH and [
3H]-CpG-mal in undiluted mouse
plasma was measured at various time points by size exclusion chromatography and gel
electrophoresis. A plot of the fraction degraded versus time is shown in Figure 3.1 and the
slope indicates the [3H]-CpG-COOH and [
3H]-CpG-mal were degraded in murine plasma at a
rate of 2.3 and 1.5 % per hr, respectively, compared to a rate of 0.15 % per hr in PBS for
[3H]-CpG-mal. Gel electrophoresis of the [
3H]-CpG-COOH showed the majority of the
material corresponds to intact [3H]-CpG-COOH with a slow appears of smaller sequences.
52
3.4.2 Plasma Pharmacokinetics
There was a dramatic difference in the plasma concentration profile of the [124
I]-CpG-
mal versus [124
I]-CpG-COOH, which can be seen in Figure 3.2. The [124
I]-CpG-COOH
rapidly falls below the limit of the detection within 7 h, whereas, the [124
I]-CpG-mal has a
higher concentration and longer half-life and never reaches background levels during the
course of the experiment. When the data are fit to a two-compartment model the half-lives
for [124
I]-CpG-COOH and [124
I]-CpG-mal were 0.3 h and 10 hr respectively. The half-life of
precomplexed [124
I]-CpG-MSA was 7 hr which is similar to [124
I]-CpG-mal and half-life of
[124
I]-MSA was 13 hr, which was longer than both of the CpG-albumin conjugates.
3.4.3 Biodistribution and Tumor Accumulation
The tumor accumulation of the different CpG ODN is shown in Figure 3.3. [124
I]-
CpG-COOH reached a maximal tumor accumulation of ~2 %ID/mL approximately 30 min
after injection and begins to quickly drop thereafter. On the other hand, [124
I]-CpG-mal
reached a maximum of ~3 %ID/mL approximately 4 h after injection and slowly decreases
which is similar to ex vivo complexed [124
I]-CpG-MSA. [124
I]-MSA had a tumor
accumulation of ~5%ID/mL approximately 4 h after injection. The tumor exposure was
increased 30 fold for [124
I]-CpG-mal compared to [124
I]-CpG-COOH.
The [124
I]-CpG-COOH is rapidly excreted by the kidneys into the urine and no other
organs appear to accumulate the drug. On the other hand, [124
I]-CpG-mal, [124
I]-CpG-MSA,
and [124
I]-MSA show widespread distribution to all organs. Post-sacrifice organ distribution,
Figure 3.4, showed the majority of 124
I radiolabel to be contained within the blood and the
tumor had the next highest concentration. The predominate mode of clearance was by the
53
kidneys and also by elimination in the feces. [124
I]-MSA appeared to be exclusively
eliminated by the kidneys.
3.4.4 4T1 Tumor Growth Study
There was no difference in initial tumor growth rate prior to surgical resection
between any of the study groups. All of the groups showed a steady increase in tumor
volume during treatment and the tumors were resected one week after the first treatment.
Due to the aggressive nature of the 4T1 cells infiltrating adjacent muscle and skin tissue,
complete surgical resection without major surgical intervention was not possible, but a
drastic reduction in primary tumor burden was achieved. Skin ulcerations which were
attributed to complications from the surgery were prevalent and necessitated euthanasia. No
difference in the survival time post-surgery was observed across the study groups, as shown
in Figure 3.5.
3.4.5 In Vitro Cytokine Release
The purpose of the in vitro activation study was to test CpG-albumin conjugates are
active to antigen presenting cells in the absence of any physiological barriers limiting access
to these cells. The murine macrophage cell line J774 was incubated with various
concentrations of different CpG treatments and ELISA against IL-6 and IL-12 was performed
on the cell supernatants to determine the extent of activation. It can be seen from the results
in Figure 3.6 that the activation was reduced for CpG-albumin compared to CpG. It is also
clear from the results that phosphodiester CpG are not able to activate the macrophages as
well as the phosphorothioate CpG.
54
3.5 Discussion
It is possible that albumin conjugation may affect the interaction between CpG ODN
and enzymatic nucleases and serve to protect them from degradation. The CpG-COOH and
CpG-mal were both moderately stable when incubated with mouse plasma in vitro and the
CpG-mal was approximately 1.5 times more stable. Since the 3’-end was modified these
CpG were not subject to rapid degradation by 3’-exonucleases that are prevalent and
responsible for the majority of degradation in plasma.(Eder, DeVine et al., 1991) Other
studies indicated that in vitro stability does not correlate to in vivo stability due to higher
levels of cellular endonucleases.(Kang, Boado et al., 1995) In this study the 124
I radiolabel is
located on the crosslinker linking the CpG to albumin. The measured 124
I radioactivity gives
no information whether the CpG is still intact. Therefore the results represent the upper limit
of how long the CpG could remain assuming that degradation of the CpG does not influence
the disposition.
The 30-fold increase in tumor accumulation of [124
I]-ODN-mal compared to [124
I]-
ODN-COOH is attributed to the increased plasma t1/2 and increased vascular permeability in
the tumor vicinity.(Matsumura and Maeda, 1986) The extended retention of [124
I]-ODN-mal
compared to [124
I]-ODN-COOH by the tumor is consistent with impaired lymphatics and
reduced diffusion of the macromolecular conjugate versus the smaller ODN (74 kDa vs 7
kDa) in the tumor microenvironment.(Matsumura and Maeda, 1986) Although, the tumor
has higher accumulation, it never appreciably exceeded the plasma concentration in our
study.
55
Albumin has historically been used as a predictor of plasma volume because it does
not readily extravasate and its distribution upon injection is limited to the intravascular
volume. The initial signals for the blood concentration are lower than would be expected and
this may be related to using the heart ROI as an approximation of the blood concentration.
The total heart ROI signal is composed of both the blood and heart tissue signals. Figure 3.4
shows the concentration of 124
I in the blood was consistently greater than heart tissue so the
majority of the heart signal comes from the blood. This should lead to two signal
distribtutions, a higher one for the blood volume and a lower one for the heart tissue.
Interestingly, a histogram of the ROI signal did not reveal these two distinction signal
distributions (data not shown).
The lack of an efficacious response despite a difference in tumor accumulation
prompted us to investigate whether albumin conjugation possibly interfered with TLR9
signaling. We found that phosphodiester CpG was not a very potent agonist of TLR9 which
is in agreement with other studies and attribute this to nuclease instability.(Krieg, MATSON
et al., 1996) Replacing the CpG phosphodiester backbone with phosphorothioate
significantly increases the activation of TLR9. Modification of the 3’-end of CpG was not
expected to alter the activation,(Kandimalla, Bhagat et al., 2002) but we noticed a distinct
difference which must related to the difference in nature of modification. Conjugation of
albumin further reduced the activation but did not totally abolish the activity. It is concluded
based on the results of the in vitro activation assay that the lack of in vivo efficacy was
related to the PO CpG not being a potent TLR9 agonist.
56
Figure 3.1. In vitro plasma stability of PO [3H]-CpG. The difference in rate of degradation in plasma versus
PBS buffer indicates enzymatic degradation. Gel electrophoresis show the majority of remaining CpG is intact
with the presence of some smaller sequences. Equal size aliquots were loaded onto the gel, the reduction in
staining intensity is due to fully degraded nucleotides running off gel.
y = 2.30x
y = 1.52x
y = 0.15x
0
2
4
6
8
10
12
14
16
18
20
0 10 20 30 40
% D
egr
ade
d
Time (h)
CpG-COOH CpG-mal CpG-mal in PBS 15 m
in
1 h
r
2 h
r
4 h
r
8h
r
24 h
r
55 h
r
57
Figure 3.2. Blood concentration of 124
I labeled CpG. CpG-COOH is rapidly cleared from the circulation
whereas the other treatments exhibit prolonged blood retention. The line represent the 2-compartment model fit
parameters found in Table 3-1. Each data point is the average of 2 animals imaged simultaneously.
0.0
0.1
1.0
10.0
100.0
0.00 10.00 20.00 30.00 40.00
He
art
Sign
al (
%ID
/ml)
time (hr)
CpG-COOH MSA CpG-MSA CpG-mal
58
Table 3-1. Plasma and tumor exposure of the [124
I]-labeled CpG.
AUC were calculated by non-compartment analysis of the respective time activity curves. Initial volume and
compartmental rate constants are calculated from a 2-compartment model.
Plasma AUC
(h %ID/ml)
Tumor AUC
(h %ID/ml)
V1
(mL)
k10
(h-1
)
k12
(h-1
)
k21
(h-1
)
[124
I]-CpG-COOH 3 ± 1 7 ± 1 14.5 ± 1.4 2.4 ± 0.3 8.8 ± 1.0 1.1 ± 0.1
[124
I]-CpG-mal 196 ± 19 (69) 200 ± 32 (29) 8.5 ± 0.2 0.07 ± 0.02 0.5 ± 0.1 1.2 ± 0.2
[124
I]-CpG-MSA 185 ± 8 (65) 194 ± 99 (29) 8 ± 1.2 0.10 ± 0.01 0.6 ± 0.2 0.4 ± 0.1
[124
I]-MSA 305 ± 30 (107) 338 ± 103 (50) 6.8 ± 0.1 0.05 ± 0.01 0.43 ± 0.03 1.0 ± .09
59
Figure 3.3. Tumor time activity curve showing tumor uptake of 124
I labeled CpG. The CpG-COOH
quickly rises and falls whereas all other treatments show increased accumulation and
retention in the tumor. Each data point is the average of 2 animals imaged simultaneously.
0
1
2
3
4
5
6
0.00 10.00 20.00 30.00 40.00
Tum
or
Sign
al (
%ID
/ml)
time (hr)
CpG-COOH MSA CpG-MSA CpG-mal
60
Figure 3.4. Terminal biodistribution of 124
I-labeled CpG measured by ex vivo gamma counting. The time post
injection the samples were extracted from the mice is shown in the legend. For MSA and CpG-mal, N=4; for
CpG-COOH and CpG-MSA, N =2. Error bars represent the standard deviation of the measurements. For CpG-
mal and MSA, one mouse had large lung uptake which was attributed to the formation of a clot, otherwise the
blood had the highest concentration followed by the tumor for all albumin associated formulations.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Blood Heart Kidney Liver Lung Spleen Tumor
Co
ne
ntr
atio
n (
%ID
/g t
issu
e)
Organ
CpG-COOH (24h) MSA (42h) CpG-MSA (31h) CpG-mal (35h)
61
Figure 3.5. Survival after surgical resection of 4T1 primary tumor. Red arrows represent treatment, the black
arrow represents surgical resection. In all treatment groups showed uniform lethality by 35 days after tumor
inoculation and mice were euthanized prematurely due to skin ulcerations. For all treatment groups N =8.
0
0.2
0.4
0.6
0.8
1
0 10 20 30 40
Surv
ival
Fra
ctio
n
Time (d)
CpG-mal
GpC-mal
CpG-COOH
PBS
CpG
62
Figure 3.6. In vitro IL-12 and IL-6 release from J444 cells. ELISA was performed on cell supernatant after 20
h of incubation.
0
5
10
15
20
25
30
0.0 0.1 1.0 10.0 100.0
IL-6
(n
g/m
L)
CpG ODN (µg/mL)
PO CpG
PO CpG-COOH
CpG
CpG-COOH
PO CpG-MSA
CpG-MSA
0
10
20
30
40
50
60
70
0.0 0.1 1.0 10.0 100.0
IL-1
2 (
ng
/mL
)
CpG ODN (µg/mL)
PO CpG
PO CpG-COOH
CpG
CpG-COOH
PO CpG-MSA
CpG-MSA
63
REFERENCES
Eder, P. S., R. J. DeVine, J. M. Dagle and J. A. Walder (1991). "Substrate specificity and
kinetics of degradation of antisense oligonucleotides by a 3' exonuclease in plasma."
Antisense Res Dev 1(2): 141-151.
Elsadek, B. and F. Kratz (2012). "Impact of albumin on drug delivery - New applications on
the horizon." J Control Release 157(1): 4-28.
Graham, M. J., S. M. Freier, R. M. Crooke, D. J. Ecker, R. N. Maslova and E. A. Lesnik
(1993). "Tritium labeling of antisense oligonucleotides by exchange with tritiated
water." Nucleic Acids Res 21(16): 3737-3743.
Kandimalla, E. R., L. Bhagat, D. Yu, Y. Cong, J. Tang and S. Agrawal (2002). "Conjugation
of ligands at the 5'-end of CpG DNA affects immunostimulatory activity." Bioconjug
Chem 13(5): 966-974.
Kang, Y. S., R. J. Boado and W. M. Pardridge (1995). "Pharmacokinetics and organ
clearance of a 3'-biotinylated, internally [32P]-labeled phosphodiester
oligodeoxynucleotide coupled to a neutral avidin/monoclonal antibody conjugate."
Drug Metab Dispos 23(1): 55-59.
Kawarada, Y., R. Ganss, N. Garbi, T. Sacher, B. Arnold and G. J. Hammerling (2001). "NK-
and CD8(+) T cell-mediated eradication of established tumors by peritumoral
injection of CpG-containing oligodeoxynucleotides." J Immunol 167(9): 5247-5253.
Kratz, F., A. Warnecke, K. Scheuermann, C. Stockmar, J. Schwab, P. Lazar, P. Druckes, N.
Esser, J. Drevs, D. Rognan, C. Bissantz, C. Hinderling, G. Folkers, I. Fichtner and C.
Unger (2002). "Probing the cysteine-34 position of endogenous serum albumin with
thiol-binding doxorubicin derivatives. Improved efficacy of an acid-sensitive
doxorubicin derivative with specific albumin-binding properties compared to that of
the parent compound." J Med Chem 45(25): 5523-5533.
Krieg, A. M., S. MATSON and E. FISHER (1996). "Oligodeoxynucleotide modifications
determine the magnitude of B cell stimulation by CpG motifs." Antisense and Nucleic
Acid Drug Development 6(2): 133-139.
Kunikata, N., K. Sano, M. Honda, K. Ishii, J. Matsunaga, R. Okuyama, K. Takahashi, H.
Watanabe, G. Tamura, H. Tagami and T. Terui (2004). "Peritumoral CpG
oligodeoxynucleotide treatment inhibits tumor growth and metastasis of B16F10
melanoma cells." J Invest Dermatol 123(2): 395-402.
64
Loening, A. M. and S. S. Gambhir (2003). "AMIDE: a free software tool for multimodality
medical image analysis." Mol Imaging 2(3): 131-137.
Ma, X. J., S. Dahiya, E. Richardson, M. Erlander and D. C. Sgroi (2009). "Gene expression
profiling of the tumor microenvironment during breast cancer progression." Breast
Cancer Res 11(1): R7.
Matsumura, Y. and H. Maeda (1986). "A new concept for macromolecular therapeutics in
cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the
antitumor agent smancs." Cancer Res 46(12 Pt 1): 6387-6392.
Palma, E. and M. J. Cho (2007). "Improved systemic pharmacokinetics, biodistribution, and
antitumor activity of CpG oligodeoxynucleotides complexed to endogenous
antibodies in vivo." J Control Release 120(1-2): 95-103.
Tlsty, T. D. and L. M. Coussens (2006). "Tumor stroma and regulation of cancer
development." Annu Rev Pathol 1: 119-150.
Verel, I., G. W. Visser, M. J. Vosjan, R. Finn, R. Boellaard and G. A. van Dongen (2004).
"High-quality 124I-labelled monoclonal antibodies for use as PET scouting agents
prior to 131I-radioimmunotherapy." Eur J Nucl Med Mol Imaging 31(12): 1645-
1652.
CHAPTER IV: PHARMACOKINETICS/BIODISTRIBUTION AND
PHARMACODYNAMICS OF MALEMIDE-DERIVATIZED
OLIGODEOXYNUCLEOTIDES WITH PHOSPHOROTHIOATE
BACKBONE IN TUMOR-BEARING MICE
4.1 Overview
In an attempt to increase the efficacy of systemically injected CpG
oligodeoxynucleotides with phosphorothioate backbone for cancer immunotherapy, they
were modified with a maleimide group (CpG-mal) to promote in situ conjugation with Cys34
of serum albumin. The CpG-mal had a slower plasma distribution and longer plasma half-
life than control CpG (CpG-COOH) that cannot form such a conjugate. The CpG-mal had
lower kidney accumulation and increased spleen and liver accumulation compared to CpG-
COOH as well as increased tumor concentrations. CpG-mal and CpG-COOH were both able
to cause complete tumor regression in 60% of the mice bearing CT26 tumors but no tumor
regression was observed for 4T1 tumors. Increasing the length of the maleimide crosslinker
or incorporating a reducible disulfide into the crosslinker did not significantly affect the in
vitro activation of macrophages by CpG-albumin conjugates.
4.2 Introduction
CpG oligodeoxynucleotides (ODN) mimic bacterial DNA and act as toll-like receptor
(TLR) 9 agonists to cause activation of the innate and adaptive immune response.(Krieg, Yi
66
et al., 1995) CpG monotherapy is effective in murine cancer models upon peritumoral
injection but never when administered systemically.(Heckelsmiller, Rall et al., 2002;
Nierkens, den Brok et al., 2009) The anti-tumor response appears to primarily involve CD8+
T cells, NK cells, and macrophages rather than CD4+ T cells.(Heckelsmiller, Rall et al.,
2002; Ballas, 2007; Buhtoiarov, Sondel et al., 2007) The lack of efficacy for systemically-
injected CpG is presumably caused by low exposure due to the rapid plasma clearance of
CpG. In a previous study in our lab, it was shown that tumor exposure and efficacy of
systemically injected CpG could be increased by making in situ monomeric IgG complexes
of hapten-derivatized CpG injected into mice preimmunized against the hapten.(Palma and
Cho, 2007) While the results were quite promising, the clinical use may be limited due to
low steady-state titers of specific IgG in humans. In search of an alternative endogenous
serum proteins that could be used as a carrier we have shown that maleimide-modified
phosphodiester CpG were able to quickly react with serum albumin. The in situ conjugation
significantly altered the biodistribution increasing both the plasma half-life and tumor
accumulation. However, these CpG were not able to activate TLR9 to produce an immune
response. The present study was designed to test whether serum albumin could serve as a
carrier of phosphorothioate CpG for cancer therapy in a similar fashion to immunoglobulins.
In other studies where albumin has been used as a carrier of cytotoxic anticancer
drugs the target has always been the cancer cells themselves. This is in contrast to our study
where the CpG-albumin target is the dendritic cells and macrophages within the tumor
periphery. Accessing the tumor periphery is a much easier task due to the heterogeneous
transport phenomena within tumor tissue.(Chauhan, Stylianopoulos et al., 2011)
67
Additionally, phagocytic cells are expected to accumulate more albumin than nonphagocytic
cells.(Chang, 1969)
4.3 Experimental Procedures
CpG Chemistry
The CpG1826 used in all experiments were purchased from Integrated DNA
Technologies (Coralville, IA) with a phosphorothioate backbone as the Na+ salt. The
sequence was TCCATGACGTTCCTGACGTT and contained a commercially available 3’
amino modification. CpG1826 with no modifications was also purchased.
HPLC Conditions
All HPLC analysis and purification was performed using Shimadzu SCL-10A system
controller with two Shimadzu LC-8A pumps connected to a Rainin Dynamax UV-C detector
and a Shimadzu C-R6A Chromatopac recorder. Solvent A was 5% acetonitrile in 100 mM
triethylammonium acetate buffer; solvent B was 100% acetonitrile. For analytical work an
Agilent Zorbax 300SB-C18 4.6 x 150mm analytical column with 5 µm particle size was used
with a total flow rate of 1 ml/min using the following gradient: t = 0-5 min, %B = 0; t = 5-30
min, %B = 0-25; t = 30-33 min, %B = 25-100. For purification work, Solvent A was 5%
acetonitrile in 10 mM triethylammonium acetate buffer and the same gradient protocol was
used with an Agilent Zorbax 300SB-C18 9.4 x 250mm semi-preparative column and a total
flow rate of 4 ml/min. All detection was performed at λ = 260 nm.
CpG-mal (Figure 4.1)
68
To a microcentrifuge tube containing CpG-NH2 in 100 mM sodium phosphate buffer
pH = 7.4 was added 20 equivalents of EMCS in acetonitrile. After 90 min at 25°C the
reaction was judged complete by analytical HPLC. The mixture was concentrated under a
stream of N2 and purified using semi-preparative HPLC. The desired peak was manually
collected and concentrated in vacuo after acidification to pH = 5.2 by the addition of excess 3
M sodium acetate buffer. The CpG-mal was ethanol precipitated from 0.3 M sodium acetate
to give the Na+ salt.
CpG-COOH (Figure 4.1)
To a microcentrifuge tube containing CpG-mal was added 50 mM NaOH solution.
After 2 h at 37°C the reaction was judged complete by analytical HPLC. The CpG-COOH
was ethanol precipitated from 0.3 M sodium acetate at -20°C to give the Na+ salt.
CpG-PEG24-mal
To a microcentrifuge tube containing CpG-NH2 in 100 mM sodium phosphate buffer
pH = 7.4 was added 20 equivalents of NHS-PEG24-mal in acetonitrile. After 90 min at 25°C
the reaction mixture separated by repeated ultracentrifugation using 5 kDa MWCO filters in
0.3 M sodium acetate buffer pH = 5.2 to remove unreacted crosslinker. The CpG-PEG24-
mal was ethanol precipitated from 0.3 M sodium acetate at -20°C to give the Na+ salt, yield:
60%.
CpG-SS-mal
To a microcentrifuge tube containing CpG-NH2 in 100 mM sodium phosphate buffer
pH = 7.4 was added 20 equivalents of S-SS-4FB (Solulink, San Diego, CA) in
69
dimethylformamide. After 90 min at 25°C the reaction mixture separated by repeated
ultracentrifugation using 5 kDa MWCO filters in 0.3 M sodium acetate pH = 5.2 buffer to
remove unreacted crosslinker and 20 equivalents of N-β-Maleimidopropionic acid hydrazide
(BMPH) was added. After 30 min at 25°C the reaction mixture was ultracentrifuged to
remove the unreacted crosslinker. The CpG-SS-mal was ethanol precipitated from 0.3 M
sodium acetate at -20°C to give the Na+ salt; yield: 55%.
[3H] Radiolabeling of CpG
CpG-NH2 was radiolabeled using the procedure of Graham, et al.(Graham, Freier et
al., 1993) Briefly, 6 mg of CpG-NH2 was lyophilized from 200 µL of 0.1 M sodium
phosphate buffer, pH = 7.8, containing 0.2 mM EDTA in microfuge tube. To the dry powder
was added 200 µL of T2O (5 Ci/g, Moravek Biochemicals; Brea, CA) and 8.3 µL of β-
mercaptoethanol. After 6h at 90°C the reaction mixture was subjected to repeated
ultracentrifugation using 5kDa MWCO filters to remove the bulk of excess T2O.
Exchangeable tritium was removed by several rounds of suspension in 1 mL H2O, incubation
at 25°C for 1 h, and lyophilization. The [3H]-CpG-NH2 (4.1 mg; 68% yield) was stored as a
dry power at -20℃ until use. The specific activity (SA) was 3.1 x 104
DPM/µg. Analytical
HPLC indicated no degradation had occurred during the procedure and the [3H]-CpG-NH2
was processed with EMCS as above.
4.3.1 In Vitro Cytokine Release
J774 cells were obtained from the UNC Tissue Culture Facility, with provenance
from ATTC (Manassas, VA), and were cultured with Dubelco modified essential media
supplemented with 10% bovine calf serum in 5% CO2 atmosphere at 37°C. The cells were
70
grown in T150 flasks and harvested by scraping at 80% confluence. All experiments were
performed with cells with a passage number less than twenty. For experiments, the entire
contents of a T-150 flask would be transferred to a 24 well plate (ca. 7 x 105 cells/well by
hemocytometer count) in 1 mL/well of media. After 4 h, the media was removed, and
replaced with 900 µL of fresh media and 100 µL of PBS containing the desired treatment.
All treatments were sterile filtered using 0.22 µm PVDF membrane filters prior to dilution
and addition to the cells. After 20 h, the media was removed and spun at 1000 x g for 5 min
at 25°C. The supernatant was carefully removed and assayed within 1 h or frozen on dry ice
and stored at -20°C until ELISA analysis.
IL-6 and IL-12 (p40) ELISA kits and appropriate buffer sets were purchased from BD
Biosciences and 96-well plates were purchased from Nunc. The assays were performed in
accordance to manufacturer’s instructions with no modifications. When dilutions were
necessary the cell culture supernatant was diluted with assay diluent. The absorbance was
read at 450 nm and was corrected with 570 nm subtraction using a Mtech plate reader. The
concentration of cytokine was then back-calculated using the appropriate standard curve and
dilution factors. The results were assayed in duplicate and are expressed as the average ±
standard deviation.
CpG-albumin Conjugation
CpG-albumin conjugates were prepared by a method previously reported in Section
2.3.2. Briefly, maleimide containing CpG was incubated with 8-fold molar excess murine
mercaptalbumin for 2 h at 25°C in PBS. The mixture was directly loaded onto Q strong
anion spin columns (Pierce, Rockford, IL) and separated using a stepwise increasing NaCl
71
gradient in 20 mM sodium phosphate pH = 7.4 buffer. Unreacted albumin was eluted using
500 mM NaCl, CpG-albumin conjugate was eluted using 1 M NaCl, and unreacted CpG was
eluted using 2 M NaCl. Repetitive ultracentrifugation using 30 kDa MWCO filters was used
to change the buffer to PBS and the conjugate was stored at 4°C not more than 48 h until use.
Mice and Cell Lines
All mice were handled in accordance with an approved protocol by UNC Institutional
Animal Care and Use Committee. Female Balb/c mice age, 8-10 weeks, were purchased
from the National Cancer Institute (Bethesda, MD). For tumor growth inhibition studies
tumor size was measured by calipers three times per week and calculated as V = ½ab2, where
a was the longest diameter and b was the length of the perpendicular diameter in mm. Mice
were allowed to survive until a maximum diameter of 2 cm was reached at which point they
were sacrificed by cervical dislocation after CO2 euthanasia. CT26 and 4T1 cell lines were
purchased from ATCC (Manassas, VA) and grown according to the manufacturer’s
recommendations.
4.3.2 Pharmacokinetic and Biodistribution Study
Fifty mice were inoculated subcutaneously in the right flank with 2.5 x 105 CT26
cells in 50 µL of Hank’s balanced buffer solution (HBSS). After 10 days the mice were
randomly divided into groups of 3 and were given tail vein injections of 100 µg of either
[3H]CpG-mal (SA = 2.1 x 10
4 DPM/µg) or [
3H]CpG-COOH (SA = 2.1 x 10
4 DPM/µg) in
100 µL of PBS. All treatments were sterile filtered using 0.22 µm PVDF filters prior to
injection. For [3H]-CpG-COOH, 3 mice were sacrificed by cardiac puncture at the following
time points: 5, 10, 15, 30, 60, 120, 240 min after intraperitoneal injection of 100 µL of 100
72
mg/mL ketamine hydrochloride. For [3H]-CpG-mal, mice were sacrificed in a similar
manner at 10 min, 1, 4, 8, 15, 24, 42 hr. Whole blood was collected into tubes containing
EDTA and the heart, liver, spleen, kidneys, lungs, tumor and injection site were rinsed with
PBS, blotted dry, and weighed in scintillation vials. Whole organs were processed for
radioactivity analysis except for the liver, which was cut into 100 mg pieces. Whole blood
was processed in triplicate 100 µL fractions. To each tissue sample was added 1 mL of
Solvable® and was incubated overnight at 25°C, and for 2 hr in a 60°C shaking water bath.
To decolorize the samples, 100 µL of 0.1M EDTA was added followed by 200 µL of 30%
H2O2 and the samples incubated for 30 min in a shaking water bath at 60°C. After cooling to
25°C, 10 mL of Ultima Gold scintillation cocktail was added to each vial and the
radioactivity was measured using a Packard Tricarb scintillation counter and reported in
DPM. Organ-specific quench curves were generated but were found to be unnecessary as no
quenching was observed; no correction for organ residual blood volume was performed.
4.3.3 CT26 Tumor Growth Study
Thirty female Balb/c mice, age 8-10 weeks, were subcutaneously inoculated with 2.5
x 105 syngeneic CT26 cells in 50 µL HBSS into the right flank. After 10 days the mice were
randomly divided into three groups and treatment was initiated. The mice received a total of
4 tail vein iv injections every 3 days. Group A (N = 9) received 100 µl of PBS, group B (N =
10) received 100 µg of CpG-COOH in 100 µl PBS, and group C (N = 10) received 100 µg of
CpG-mal in 100 µl of PBS.
73
4.3.4 4T1 Tumor Growth Study
Forty mice were orthotopically inoculated with 1 x 104 4T1 cells in 50 µL PBS into
the mammary fat pad. After 10 days, the mice were randomly divided into three groups of
10 and experimental treatments were initiated. A total of three treatments were given every
three days. Group A received 100 µg of CpG in 100 µL PBS via tail vein, group B received
50 µg of CpG in 50 µL PBS peritumorally, and group C received 100 µg equivalent of CpG-
mal in 100 µL PBS via tail vein. All treatments were sterile filtered through a 0.22 µm
PVDF filter immediately prior to injection.
4.4 Results
4.4.1 In Vitro Macrophage Activation
It was shown previously that CpG-albumin conjugates display lower IL-12 and IL-6
release from a macrophage cell line than unconjugated control CpG (Figure 3.6). In order to
test whether steric hindrance was inhibiting the CpG-albumin from interacting with TLR9,
CpG was derivatized with a longer crosslinker containing 24 repeating ethylene glycol units,
shown in Figure 4.2. This chemical manipulation should extend the CpG moiety away from
the albumin. As shown in Figure 4.2 this modification did not increase the activation to any
appreciable extent. Additionally, in order to test whether releasing the CpG from albumin in
the reductive environment of the endosome could increase the activity, a reversible disulfide
crosslinker was employed to create a reducible linkage between CpG and albumin. The
disulfide was able to be reductively cleaved by dithiothreitol, indicating its instability in a
reductive environment (data not shown). However, the CpG-SS-albumin was still unable to
increase the IL-12 or IL-6 secretion.
74
4.4.2 Pharmacokinetics and Biodistribution
The maleimide modification was able to alter the plasma pharmacokinetics of CpG;
as expected the [3H]-CpG-COOH that cannot form an adduct with albumin was rapidly
cleared from the circulation and distributed into tissues whereas the [3H]-CpG-mal was more
slowly distributed as shown in Figure 4.3. Both of the CpG show biphasic plasma profiles
indicating distribution into tissue compartments, most likely due to PS ODN have binding to
numerous cell membrane proteins.(Beltinger, Saragovi et al., 1995)
The biodistribution of the [3H]-CpG-mal was different than [
3H]-CpG-COOH and can
be seen in Figure 4.5. The major difference was found in the liver and kidney. For [3H]-
CpG-COOH the kidney had the highest accumulation of drug, whereas for [3H]-CpG-mal the
liver had the highest accumulation. The spleen also had higher accumulation for [3H]-CpG-
mal than [3H]-CpG-COOH. The tumor accumulation of [
3H]-CpG-mal is presented in
Figure 4.4 and was slightly increased compared to [3H]-CpG-COOH. There is a clear
downward trend for [3H]-CpG-COOH tumor concentration from 1-4 h, whereas the [
3H]-
CpG-mal is increasing during this time.
4.4.3 Tumor Growth Studies
In the CT26 model, a dramatic difference was seen between the groups receiving
CpG therapy and the PBS group, and little difference was observed between the CpG-COOH
and CpG-mal groups. The mice treated with CpG-COOH and CpG-mal showed a slow
increase in tumor size during treatment followed by a complete regression in tumor volume.
When the individual mice tumor volumes are examined in Figure 4.8 it is evident that there
are clear-cut responders and non-responders to the CpG monotherapy. There also appears to
75
be slightly more variability in the response to the CpG-COOH compared to CpG-mal. On
the other hand, in Figure 4.6 the 4T1 model the tumor growth was only slowed during
treatment, and when treatment was stopped the tumors quickly began to grow. Additionally,
severe toxicity was observed after the second treatment for both of the systemically-injected
treatments in this model.
4.5 Discussion
Figure 4.2 demonstrates CpG-albumin conjugates are able to activate the immune
system, presumably via TLR9, albeit to a weaker extent than free CpG. The length of the
crosslinker did not have a significant effect on the in vitro activation of macrophages,
suggesting steric hindrance by the conjugated albumin does not interfere with TLR9
signaling. Incorporation of a reversible disulfide bond should generate free CpG if the bond
is reduced. This should increase the activation of the macrophages to levels comparable to
free CpG. The observation that the reducible disulfide crosslinker was not able to increase
the in vitro activation is most likely explained by the crosslinker not being biologically
reduced during the timeframe of TLR9 signaling. This is in contrast to a study that linked
CpG to an antibody using a disulfide linkage and found that a cleavable linker was needed
for immune cell activation. It should be noted that study used a 5’ conjugated CpG which is
known to reduce the activation(Kandimalla, Bhagat et al., 2002), additionally, in their in vitro
incubation experiments the media was supplemented with β-mercaptoethanol, which could
have provided an artificial reductive environment.(Sharma, Dominguez et al., 2008)
Whether our observation can be extended to other phagocytic cell types, or if the in vivo
redox state is faithfully mimicked in vitro, remains to be seen. However, if so, it suggests
that reduction would need to occur prior to internalization. The difference in activation may
76
well be due to a difference in the internalization mechanism between the CpG-albumin
versus free CpG.
Our results are in agreement with the observation that the maleimide group is
compatible with phoshpohthiote ODN.(Sanchez, Pedroso et al., 2012) The plasma half-life
of CpG ODN was altered by the conjugation of a maleimide group which allowed the CpG-
mal to covalently react with circulating albumin. Since the plasma clearance represents
distribution rather than elimination the CpG-mal which has distributed out of the vasculature
prior to albumin conjugation may still react with the interstitial albumin. After conjugation,
the CpG-albumin displayed a slower plasma distribution, as would be predicted with an
increase in hydrodynamic size. Also consistent with this finding is the observation that the
kidney accumulation was reduced for the CpG-mal compared to CpG-COOH. The addition
of the 3’ amino modification and subsequent conjugation with EMCS is not expected to
dramatically alter the distribution of CpG. In another study where a PS ODN was modified
with a 5’ octadecyl-amine moiety, the distribution and clearance was marginally
altered.(Crooke, Graham et al., 1996)
Interestingly, the liver accumulation of CpG-mal was significantly increased
compared to CpG-COOH. This may be due to the fact that the plasma half-life was
increased, promoting more interaction with richly-perfused liver tissue. Since the liver
accumulation was not rapid, it does not appear to involve high-affinity scavenger receptors
that are thought to specifically bind degraded albumin.(Schnitzer and Bravo, 1993) The liver
uptake is not expected to be saturated at the doses of CpG that were administered in this
study.(Graham, Crooke et al., 1998) Also, the large amount of liver uptake appears to be
inherent to phosphorothioate ODN as it is not seen with phosphodiester ODN.(Sands, Gorey-
77
Feret et al., 1994) The slight increased spleen uptake could lead to an increase in efficacy
due to the high amount of antigen presenting cells in this organ. Alternatively, the increased
uptake by the spleen and liver could lead to non-specific systemic immune response which
may have caused the increase in toxicity observed in the 4T1 study.(Martin-Armas, Simon-
Santamaria et al., 2006)
The difference in tumor accumulation was less striking than would be predicted based
on the difference in plasma concentration. Again, we attribute this to the high liver clearance
of the CpG-mal. Additionally, CpG-mal had a slightly increased tumor accumulation
compared to CpG-COOH and it appears the tumor retention was also increased. This increase
in exposure may account for the less variable response of the CpG-mal compared to CpG-
COOH in the CT26 tumor model.
The efficacy of CpG therapy was found to be dependent on the tumor model. In the
CT26 model tumor regression was observed, whereas no regression was seen in the 4T1
model. Since CpG is not inherently cytotoxic, the only way for there to be delayed
regression in tumor volume would be if an adaptive immune response involving mostly
CD8+ cytotoxic lymphocytes against the tumor was generated. The fact that there was
observed regression in the CT26 model suggests a specific immune response was generated.
This is supported by the finding that the CT26 model is known to be moderately
immunogenic(Belnap, Cleveland et al., 1979), whereas the 4T1 model is non-
immunogenic.(Pulaski and Ostrand-Rosenberg, 1998) Only studies which tested a
combination of CpG with an apoptosis-inducing component were found to be successful in
the 4T1 model.(Mroz, Castano et al., 2009; Shirota and Klinman, 2011) Clearly the need for
78
tumor-associated antigens is critical for CpG therapy, as well as for all adjuvant
immunotherapies in general.
It has been reported that tumor burden when CpG therapy is initiated determines the
efficacy, with small tumors showing a better response.(Heckelsmiller, Rall et al., 2002;
Sharma, Karakousis et al., 2004) The tumor size in the current study was intermediate
compared to these previous studies but there was no correlation in the current study between
the initial size at treatment and the non-responders. Unexpectedly, CpG-COOH was also
found to be efficacious when administered systemically. This may be explained by the
dosing schedule used in the current study, administering the maximally-tolerated dose.
Alternatively, this may be due to the differences in the activation of immune cells by CpG-
COOH versus CpG. It is also interesting to note the difference in individual tumor growth
patterns. For the CpG-COOH, it appears the tumors had more variability in growth during
treatment, whereas the CpG-mal had reliably more controlled growth. This suggests there is
a difference in the immediate cytokine release and NK cell activation. Minimizing the tumor
growth during treatment is clinically important especially if the comorbidities are associated
with the tumor size.
79
Figure 4.1. Chemical structures and synthetic scheme for the
synthesis of CpG-mal and CpG-COOH.
80
Figure 4.2. In vitro IL-12 and IL-6 release from CpG-albumin conjugates.The chemical structures of the different
CpG-mal to prepare CpG-albumin conjugates are shown. It does not appear that the activtion of the macrophages is
sensitive to the crosslinker.
CpG-SS-MSA
CpG-24-MSA
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
0.0 0.1 1.0 10.0 100.0
IL-1
2 (
ng
/mL
)
CpG ODN (µg/mL)
CpG
CpG-COOH
CpG-MSA
CpG-24-MSA
CpG-SS-MSA
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
0.0 0.1 1.0 10.0 100.0
IL-6
(n
g/m
L)
CpG ODN (µg/mL)
CpG
CpG-COOH
CpG-MSA
CpG-24-MSA
CpG-SS-MSA
81
Figure 4.3. Blood concentration of [3H]-CpG. The CpG-
mal exhibits much less initial distribution than CpG-
COOH. The lines represent the 2-compartment model
parameters in Table 4-1 fit to the data. For each
timepoint 3 mice were sacrificed. The error bars
represent the standard deviation.
0.1
1.0
10.0
100.0
0 10 20 30 40
Blo
od
Co
ncen
trati
on
(µ
g/g
)
Time (h)
[3H]-CpG-COOH
82
Table 4-1. Two-compartment model parameters for [3H]-labeled CpG.
V1
(mL)
k10
(h-1
)
k12
(h-1
)
k21
(h-1
)
[3H]-CpG-COOH 3.8 ± 1.4 4.0 ± 1.5 5.6 ± 1.0 1.5 ± 0.8
[3H]-CpG-mal 2.7 ± 0.3 0.3 ± 0.02 0.9 ± 0.2 0.08 ± 0.04
83
Figure 4.4. Tumor accumulation of PS [3H]-CpG. The
tumor accumulation appears to quickly saturate for both
treatments. There is a clear downward trend for CpG-
COOH between 1-4 h whereas, the CpG-mal is increasing
during this time. . For each timepoint 3 mice were
sacrificed. The error bars represent the standard deviation.
0
2
4
6
8
10
0 10 20 30 40
Tum
or
Co
nce
ntr
atio
n (
µg/
g ti
ssu
e)
Time (h)
CpG-COOHCpG-MAL
84
Figure 4.5. Biodistribution of [3H]-labeled CpG. For the
CpG-COOH, each bar represents data from each of the
following timepoints: 5, 10, 15, 30, 60, 120, 240 min. For
CpG-mal, each bar represents data for the following
timepoints: 10 min, 1, 4, 8, 15, 24, 40 h. Each timepoint
is the average concentration calculated from 3 mice and
the error bars represent the standard deviation.
0
10
20
30
40
50
60
Co
ncen
trati
on
(µ
g/g
tis
su
e)
CpG-COOH
0
10
20
30
40
50
60
Co
ncen
trati
on
(µ
g/g
tis
su
e)
CpG-mal
85
Figure 4.6. Tumor growth inhibition of 4T1 tumors.
The growth appears to be slowed during treatment but
promptly resumes after the last dose. N=4 for CpG (iv);
N=10 for CpG (pt); N=3 for CpG-mal (iv); and N = 7 for
untreated. Severe toxicity was observed in this model for
both systemically injected treatments.
0
100
200
300
400
500
0 5 10 15 20 25
Tu
mo
r V
olu
me
(m
m3)
Time (d)
CpG (iv)CpG (pt)CpG-mal (iv)Untreated
86
Figure 4.7. CT26 tumor growth curves. Both CpG-
COOH and CpG-mal caused marked tumor growth
reduction compared to PBS. Four total treatments were
given starting day 10 every 3 days.
0
500
1000
1500
2000
2500
3000
0 10 20 30 40 50
Av
era
ge
Tu
mo
r V
olu
me
(m
m3)
Time (d)
PBS
CpG-COOH
CpG-MAL
87
Figure 4.8. Individual growth curves for CT26 tumors.
There were clear responders and non-responders to both
treatments. The CpG-mal has less variability in growth
during treatment. The arrows represent treatments.
0
100
200
300
400
500
600
700
800
0 10 20 30 40 50
Tum
or
Vo
lum
e (m
m3)
Time (d)
CpG-COOH
CpG-mal
88
REFERENCES
Ballas, Z. K. (2007). "Modulation of NK cell activity by CpG oligodeoxynucleotides."
Immunol Res 39(1-3): 15-21.
Belnap, L. P., P. H. Cleveland, M. E. Colmerauer, R. M. Barone and Y. H. Pilch (1979).
"Immunogenicity of chemically induced murine colon cancers." Cancer Res 39(4):
1174-1179.
Beltinger, C., H. Saragovi, R. Smith, L. LeSauteur, N. Shah, L. DeDionisio, L. Christensen,
A. Raible, L. Jarett and A. Gewirtz (1995). "Binding, uptake, and intracellular
trafficking of phosphorothioate-modified oligodeoxynucleotides." Journal of Clinical
Investigation 95(4): 1814.
Buhtoiarov, I. N., P. M. Sondel, J. C. Eickhoff and A. L. Rakhmilevich (2007).
"Macrophages are essential for antitumour effects against weakly immunogenic
murine tumours induced by class B CpG-oligodeoxynucleotides." Immunology
120(3): 412-423.
Chang, Y.-H. (1969). "Studies on phagocytosis: I. Uptake of radio-iodinated (131I) human
serum albumin as a measure of the degree of phagocytosis in vitro." Exp Cell Res
54(1): 42-48.
Chauhan, V. P., T. Stylianopoulos, Y. Boucher and R. K. Jain (2011). "Delivery of molecular
and nanoscale medicine to tumors: transport barriers and strategies." Annu Rev Chem
Biomol Eng 2: 281-298.
Crooke, S. T., M. J. Graham, J. E. Zuckerman, D. Brooks, B. S. Conklin, L. L. Cummins, M.
J. Greig, C. J. Guinosso, D. Kornbrust, M. Manoharan, H. M. Sasmor, T. Schleich, K.
L. Tivel and R. H. Griffey (1996). "Pharmacokinetic properties of several novel
oligonucleotide analogs in mice." Journal of Pharmacology and Experimental
Therapeutics 277(2): 923-937.
Graham, M. J., S. T. Crooke, D. K. Monteith, S. R. Cooper, K. M. Lemonidis, K. K. Stecker,
M. J. Martin and R. M. Crooke (1998). "In vivo distribution and metabolism of a
phosphorothioate oligonucleotide within rat liver after intravenous administration."
Journal of Pharmacology and Experimental Therapeutics 286(1): 447-458.
Graham, M. J., S. M. Freier, R. M. Crooke, D. J. Ecker, R. N. Maslova and E. A. Lesnik
(1993). "Tritium labeling of antisense oligonucleotides by exchange with tritiated
water." Nucleic Acids Res 21(16): 3737-3743.
89
Heckelsmiller, K., K. Rall, S. Beck, A. Schlamp, J. Seiderer, B. Jahrsdorfer, A. Krug, S.
Rothenfusser, S. Endres and G. Hartmann (2002). "Peritumoral CpG DNA elicits a
coordinated response of CD8 T cells and innate effectors to cure established tumors
in a murine colon carcinoma model." J Immunol 169(7): 3892-3899.
Kandimalla, E. R., L. Bhagat, D. Yu, Y. Cong, J. Tang and S. Agrawal (2002). "Conjugation
of ligands at the 5'-end of CpG DNA affects immunostimulatory activity." Bioconjug
Chem 13(5): 966-974.
Krieg, A. M., A. K. Yi, S. Matson, T. J. Waldschmidt, G. A. Bishop, R. Teasdale, G. A.
Koretzky and D. M. Klinman (1995). "CpG motifs in bacterial DNA trigger direct B-
cell activation." Nature 374(6522): 546-549.
Martin-Armas, M., J. Simon-Santamaria, I. Pettersen, U. Moens, B. Smedsrod and B.
Sveinbjornsson (2006). "Toll-like receptor 9 (TLR9) is present in murine liver
sinusoidal endothelial cells (LSECs) and mediates the effect of CpG-
oligonucleotides." J Hepatol 44(5): 939-946.
Mroz, P., A. P. Castano and M. R. Hamblin (2009). "Stimulation of dendritic cells enhances
immune response after photodynamic therapy." 717803-717803.
Nierkens, S., M. H. den Brok, T. Roelofsen, J. A. Wagenaars, C. G. Figdor, T. J. Ruers and
G. J. Adema (2009). "Route of administration of the TLR9 agonist CpG critically
determines the efficacy of cancer immunotherapy in mice." PLoS One 4(12): e8368.
Palma, E. and M. J. Cho (2007). "Improved systemic pharmacokinetics, biodistribution, and
antitumor activity of CpG oligodeoxynucleotides complexed to endogenous
antibodies in vivo." J Control Release 120(1-2): 95-103.
Pulaski, B. A. and S. Ostrand-Rosenberg (1998). "Reduction of established spontaneous
mammary carcinoma metastases following immunotherapy with major
histocompatibility complex class II and B7.1 cell-based tumor vaccines." Cancer Res
58(7): 1486-1493.
Sanchez, A., E. Pedroso and A. Grandas (2012). "Conjugation reactions involving
maleimides and phosphorothioate oligonucleotides." Bioconjug Chem 23(2): 300-
307.
Sands, H., L. J. Gorey-Feret, A. J. Cocuzza, F. W. Hobbs, D. Chidester and G. L. Trainor
(1994). "Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I.
Comparison of a phosphodiester and a phosphorothioate." Mol Pharmacol 45(5): 932-
943.
Schnitzer, J. E. and J. Bravo (1993). "High affinity binding, endocytosis, and degradation of
conformationally modified albumins. Potential role of gp30 and gp18 as novel
scavenger receptors." J Biol Chem 268(10): 7562-7570.
90
Sharma, S., A. L. Dominguez, S. Z. Manrique, F. Cavallo, S. Sakaguchi and J. Lustgarten
(2008). "Systemic targeting of CpG-ODN to the tumor microenvironment with anti-
neu-CpG hybrid molecule and T regulatory cell depletion induces memory responses
in BALB-neuT tolerant mice." Cancer Res 68(18): 7530-7540.
Sharma, S., C. P. Karakousis, H. Takita, K. Shin and S. P. Brooks (2004). "Cytokines and
chemokines are expressed at different levels in small and large murine colon-26
tumors following intratumoral injections of CpG ODN." Neoplasia 6(5): 523-528.
Shirota, H. and D. M. Klinman (2011). "CpG-conjugated apoptotic tumor cells elicit potent
tumor-specific immunity." Cancer Immunol Immunother 60(5): 659-669.
CHAPTER V: CONCLUSIONS & FUTURE DIRECTION
The goal of this dissertation was to determine if serum albumin had potential utility as
a carrier of CpG for cancer therapy. The collective results support the notion that albumin
may have a utility, albeit limited, in this narrowly-defined scope.
CpG oligodeoxynucleotide (ODN) was modified with maleimide groups to allow an
in situ reaction with Cys34 of serum albumin. For phosphodiester (PO) CpG, PET/CT
imaging was used to determine the pharmacokinetics and biodistribution (PK/BD) of 124
I-
labeled CpG-mal. The in situ reaction with albumin was determined to be complete within
minutes and was sufficient to compete with the rapid plasma clearance in vivo. The plasma
half-life and tumor accumulation were increased compared to control CpG-COOH that is
unable to react with albumin. Despite this increase in tumor exposure, the PO CpG-mal was
not efficacious in the 4T1 cancer model at inhibiting tumor growth, increasing survival, or
decreasing the number of lung metastasis. Subsequent in vitro activation of macrophages
indicated that PO CpG is not a potent agonist of TLR9 which was attributed instability to
endogenous nucleases. Since the amount of cellular uptake was not directly measured in the
in vitro experiments it is unknown if a difference in uptake could account for the reduction in
cytokine release. Further experiments should be done to elucidate if any differences in in
vitro uptake is observed.
92
In contrast to PO CpG, the PS CpG was able to significantly increase the TLR9
activation and subsequent release of IL-6 and IL-12 in vitro. Conjugation of the PS CpG to
albumin reduced, but did not abolish, the activation. The reduction in activation is not
believed to be caused by steric hindrance of albumin because increasing the crosslinker
length did not increase the activation. Additionally, adding a reducible disulfide linker did
not increase the activation, which suggests biological reduction did not happen within the
timeframe of TLR9 signaling. This may support the notion that endosomes are not
sufficiently reductive environments for reducing disulfide bonds.
Tumor growth inhibition studies using the 4T1 model and PS CpG showed a slowing
of tumor growth during treatment, but after treatment was stopped the tumor growth
resumed. Additionally, severe toxicity was seen in this aggressive model which was
attributed to an additive effect of the metastatic spread and the induced inflammatory
response. The lack of efficacy in this model was attributed to the lack of shedding of tumor
antigens as the 4T1 is known to be inherently non-immunogenic.(Pulaski and Ostrand-
Rosenberg, 1998)
The physicochemical properties of PS CpG are different than PO CpG and
extrapolation of the PK/BD results for the PO CpG to PS CpG is tenuous. Attempts to
radiolabel the PS CpG with 124
I were not successful, therefore a tritium label was used to
monitor the PK/BD of PS [3H]-CpG-mal. The plasma half-life of the PS CpG-mal was
increased compared to control CpG, but the increase was not as striking as for PO CpG. This
is due to the high liver accumulation of PS CpG-mal which was not observed for PO CpG-
mal. The accumulation in the liver was not immediate, but sustained, and reached
approximately 50 µg/g tissue at 4 h. This rules out high-affinity scavenger receptors which
93
quickly remove highly modified albumin, but it may indicate low-affinity scavenger
receptors that can remove partially-modified albumin.(Stehle, Sinn et al., 1997) In support of
this is the fact that highly anionic substrates are usually taken up by scavenger receptors
located on macrophages and liver sinusoidal endothelial cells.(Yamasaki, Hisazumi et al.,
2003) Conjugation of the CpG to albumin is expected to roughly double the amount of
negative charge, rendering the conjugate a better substrate for these receptors. Although,
conjugation of PO CpG is expected to also increase the negative charge.
Unfortunately, this high liver accumulation reduces the tumor exposure. Future
studies should investigate the nature of this uptake. Is it due to receptors specific for PS
ODN, or is it due to albumin scavenger receptors recognizing the PS CpG-albumin conjugate
as degraded albumin? This could be tested by adding a competitor of these receptors and
seeing if the uptake is reduced. However, both receptors share an overlap in inhibitors, so a
specific inhibitor without a monoclonal antibody would be difficult to employ. Even so,
knowing the mechanism of liver uptake does not materially change the fact that it happens
and it is a significant limitation for the therapeutic potential of PS CpG. If the uptake is
determined to be caused by receptors recognizing the PS CpG, then reducing the length of
the CpG may reduce the affinity towards these receptors, although this may also reduce the
TLR9 affinity as well. Another option is to use chimeric CpG which contains both PS and
PO backbone which should reduce the non-specific interactions caused by the PS. But this
also may have nuclease instability due to the presence of endonucleases that can cleave the
PO bonds.(Boado, Kang et al., 1995) It may also be possible for future CpG chimeras that
have other backbone modifications to be used as well.
94
Interestingly, the tumor appears to quickly accumulate PS CpG, which may be due to
some type of binding process to cellular targets.(Beltinger, Saragovi et al., 1995) The tumor
concentration for CpG-mal also quickly reaches close to maximal level, but still shows an
increase in tumor accumulation with time. This increase stems from the increased half-life of
the CpG-albumin conjugate in the circulation. In both biodistribution experiments the tumor
concentration reached a maximal value of approximately 3-6 %ID/g tissue. Since the
average mice body weight used in the studies was approximately 18-20 g, it appears that on a
per-weight basis selective accumulation in the tumor was minimal.
The antitumor effects of CpG-mal and CpG-COOH were clearly observed in the
CT26 colon cancer model. Both treatments showed a remarkable reduction in growth rate
during treatment, and between 6-10 days after the first treatment the tumors began to regress.
This timeline is consistent with the expected activation and infiltration of CD8+ T cells.
Examination of the individual tumor growth curves suggest 6/10 mice had full regression of
the tumors while the others had stasis or partial regression following by rapid regrowth of the
tumors suggesting a lack of an adaptive immune response. These results are typical of nearly
all immunotherapies and were even observed in Dr. Coley’s initial experiments.
The increased retention by the tumor of CpG-mal compared to CpG-COOH suggests
that decreasing the dose or dosing frequency may magnify the difference in the efficacy
between the two treatments. This could be tested in future dose-response studies, where a
clinical response may still be observed with less frequent or lower dose injections for CpG-
mal but not CpG-COOH.
95
The immune response in the tumor microenvironment is delicate and sometimes
activation of the immune system can increase the growth rate of tumors and increase the
chance of metastasis.(DeNardo, Barreto et al., 2009) Therefore, the manner in which the
immune system is manipulated in the tumor microevironment is critically important. The
fact that we noticed different cytokine release patterns for CpG-albumin compared to free
CpG could have a dramatic effect on the nature of the local immune response. This is
indicated by the more controlled growth observed in Figure 4.8. In future studies, a wider
variety of cytokines could be measured to determine the differences in activation between the
CpG-albumin and free CpG.
The tumor growth inhibition assay, while clinically important, is a crude measure of
efficacy. Cellular mechanistic studies should be investigated in future experiments. The
most important measure of efficacy is the infiltration of tumor-specific CD8+ T cells. Flow
cytometry could be performed on single cells suspensions of excised tumors to determine the
percentage of CD8+ T cells in the tumor. Additionally, dendritic cells from the tumor-
draining lymph nodes could be cultured and stimulated with tumor homogenates and the
expression of activation markers could be quantified by flow cytometry.
Long-term toxicity was not assayed during the current studies, but it does not appear
that conjugation of the CpG to albumin caused a catastrophic immune response to albumin.
The fact that CpG-albumin is present in the circulation for longer times in an active form is
not inherently a problem. It is known that circulating monocytes and dendritic cells have a
reduced phagocytic rate compared to peripheral tissue phagocytic cells. This should reduce
their uptake of the active CpG-albumin conjugate.
96
CpG has been conjugated to antigens in order to increase accumulation of the CpG
and the antigen in the same antigen presenting cell in order to encourage cross-
presentation.(Maurer, Heit et al., 2002) In our case though, albumin is not an antigen and an
immune response to albumin would be highly undesirable. Many studies have synthesized
albumin fusion proteins with various cytokines and have not seen deleterious immune
responses against albumin in preclinical settings; their clinical safety is under
investigation.(Chang, Gupta et al., 2009) Conceptually, it is more probable that if an
immune response were to be generated it would be against the crosslinker or the CpG rather
than native albumin. Since the PK experiments were only determined for the first dose, the
results would not be expected to be influenced by a generated immune response. In order to
test this, PK after multiple dosing should be performed. If an immune response is generated
against the CpG-albumin conjugate, then the clearance should be expected to be increased.
Since maleimide is not selective to which thiols they react with, any exposed thiols
are subject to reaction. N-ethylmaleimide (NEM) has been used to study enzyme inhibition
and has been shown to be toxic.(Schoonen, De Roos et al., 2005) However, NEM is
considerably smaller and more hydrophobic so it readily diffuses into cell membranes and
has access intracellular proteins and enzymes with catalytic thiol residues. The CpG-mal is
not expected to readily diffuse through plasma membranes and will not have access to these
intracellular thiols. This restricted distribution should dramatically reduce the maleimide-
related toxicity.
One potential drawback of using endogenous albumin as a carrier for anticancer drugs
is that cancer patients have reduced concentrations of serum albumin. This limits the amount
of carrier protein present at any given time. Additionally, during an acute immune response,
97
the transcription of albumin is known to be transiently reduced.(Liao, Jefferson et al., 1986)
This may limit the amount of albumin available for future dosing. Nevertheless, albumin
should be in a 10 fold excess compared to CpG-mal at the administered concentration.
Conjugation to albumin does not appear to abolish interaction with extracellular or
endosomal targets based off the observations that CpG-albumin was degraded by plasma
nucleases and able to activate macrophages by TLR9 signaling. This suggests that albumin
could be used as a carrier for any therapeutic whose target is accessible to native albumin,
which includes any surface, extracellular, or endosomal targets but would not include any
intracellular or nuclear targets. This no doubt limits the therapeutic opportunities but still
leaves significant number of targets which could be improved by albumin conjugation. In
summary, the use of albumin as a carrier for CpG may be limited, but that does not rule out
other potential therapeutics that could benefit from albumin conjugation.
98
REFERENCES
Beltinger, C., H. U. Saragovi, R. M. Smith, L. LeSauteur, N. Shah, L. DeDionisio, L.
Christensen, A. Raible, L. Jarett and A. M. Gewirtz (1995). "Binding, uptake, and
intracellular trafficking of phosphorothioate-modified oligodeoxynucleotides." J Clin
Invest 95(4): 1814-1823.
Boado, R. J., Y. S. Kang, D. Wu and W. M. Pardridge (1995). "Rapid plasma clearance and
metabolism in vivo of a phosphorothioate oligodeoxynucleotide with a single,
internal phosphodiester bond." Drug Metab Dispos 23(11): 1297-1300.
Chang, C.-H., P. Gupta and D. M. Goldenberg (2009). "Advances and challenges in
developing cytokine fusion proteins as improved therapeutics."
DeNardo, D. G., J. B. Barreto, P. Andreu, L. Vasquez, D. Tawfik, N. Kolhatkar and L. M.
Coussens (2009). "CD4(+) T cells regulate pulmonary metastasis of mammary
carcinomas by enhancing protumor properties of macrophages." Cancer Cell 16(2):
91-102.
Liao, W. S., L. S. Jefferson and J. M. Taylor (1986). "Changes in plasma albumin
concentration, synthesis rate, and mRNA level during acute inflammation." Am J
Physiol 251(6 Pt 1): C928-934.
Maurer, T., A. Heit, H. Hochrein, F. Ampenberger, M. O'Keeffe, S. Bauer, G. B. Lipford, R.
M. Vabulas and H. Wagner (2002). "CpG-DNA aided cross-presentation of soluble
antigens by dendritic cells." Eur J Immunol 32(8): 2356-2364.
Pulaski, B. A. and S. Ostrand-Rosenberg (1998). "Reduction of established spontaneous
mammary carcinoma metastases following immunotherapy with major
histocompatibility complex class II and B7.1 cell-based tumor vaccines." Cancer Res
58(7): 1486-1493.
Schoonen, W. G., J. A. De Roos, W. Westerink and E. Débiton (2005). "Cytotoxic effects of
110 reference compounds on HepG2 cells and for 60 compounds on HeLa, ECC-1
and CHO cells.: II Mechanistic assays on NAD (P) H, ATP and DNA contents."
Toxicology in vitro 19(4): 491-503.
Stehle, G., H. Sinn, A. Wunder, H. H. Schrenk, S. Schutt, W. Maier-Borst and D. L. Heene
(1997). "The loading rate determines tumor targeting properties of methotrexate-
albumin conjugates in rats." Anticancer Drugs 8(7): 677-685.
99
Yamasaki, Y., J. Hisazumi, K. Yamaoka and Y. Takakura (2003). "Efficient scavenger
receptor-mediated hepatic targeting of proteins by introduction of negative charges on
the proteins by aconitylation: the influence of charge density and size of the proteins
molecules." European Journal of Pharmaceutical Sciences 18(5): 305-312.
APPENDIX A
Included in the appendix are the NMR spectra and m/s analysis for the crosslinker
synthesized in Chapter II.
101
Figure A.1. 1H NMR spectrum of Mal-Tyr(tBu)-OtBu (1)
102
Figure A.2. 13
C NMR of Mal-Tyr(tBu)-OtBu (1)
103
Figure A.3. 1H NMR of Mal-Tyr (2)
104
Figure A.4. 13
C NMR of Mal-Tyr (2)
105
Figure A.5. Mass Spectrum of Mal-Tyr (2)
106
Figure A.6. 1H NMR spectrum of Mal-Tyr-TEG-COOH (3)
107
Figure A.7. 13
C NMR spectrum of Mal-Tyr-TEG-COOH (3)
108
Figure A.8. Mass spectrum of Mal-Tyr-TEG-COOH (3)
109
Figure A.9. 1H NMR of Mal-Tyr-TEG-NHS (4)
110
Figure A.10. 13
C NMR of Mal-Tyr-TEG-NHS (4)
111
Figure A.11. Mass spectrum of Mal-Tyr-TEG-NHS (4)
112
Figure A.12. Deconvoluted mass spectrum of ODN-mal.
REFERENCES
Akilesh, S., G. J. Christianson, D. C. Roopenian and A. S. Shaw (2007). "Neonatal FcR
expression in bone marrow-derived cells functions to protect serum IgG from
catabolism." J Immunol 179(7): 4580-4588.
Allison, A. C. (1960). "Turnovers of erythrocytes and plasma proteins in mammals." Nature
188: 37-40.
Andersen, J. T. and I. Sandlie (2009). "The versatile MHC class I-related FcRn protects IgG
and albumin from degradation: implications for development of new diagnostics and
therapeutics." Drug Metab Pharmacokinet 24(4): 318-332.
Andersen, J. T., J. Dee Qian and I. Sandlie (2006). "The conserved histidine 166 residue of
the human neonatal Fc receptor heavy chain is critical for the pH-dependent binding
to albumin." Eur J Immunol 36(11): 3044-3051.
Bading, J. R., M. Horling, L. E. Williams, D. Colcher, A. Raubitschek and S. E. Strand
(2008). "Quantitative serial imaging of an 124I anti-CEA monoclonal antibody in
tumor-bearing mice." Cancer Biother Radiopharm 23(4): 399-409.
Bae, Y. H. and K. Park (2011). "Targeted drug delivery to tumors: myths, reality and
possibility." J Control Release 153(3): 198-205.
Ballas, Z. K. (2007). "Modulation of NK cell activity by CpG oligodeoxynucleotides."
Immunol Res 39(1-3): 15-21.
Belnap, L. P., P. H. Cleveland, M. E. Colmerauer, R. M. Barone and Y. H. Pilch (1979).
"Immunogenicity of chemically induced murine colon cancers." Cancer Res 39(4):
1174-1179.
Beltinger, C., H. U. Saragovi, R. M. Smith, L. LeSauteur, N. Shah, L. DeDionisio, L.
Christensen, A. Raible, L. Jarett and A. M. Gewirtz (1995). "Binding, uptake, and
intracellular trafficking of phosphorothioate-modified oligodeoxynucleotides." J Clin
Invest 95(4): 1814-1823.
Bird, A. P. (1986). "CpG-rich islands and the function of DNA methylation." Nature
321(6067): 209-213.
114
Boado, R. J., Y. S. Kang, D. Wu and W. M. Pardridge (1995). "Rapid plasma clearance and
metabolism in vivo of a phosphorothioate oligodeoxynucleotide with a single,
internal phosphodiester bond." Drug Metab Dispos 23(11): 1297-1300.
Buhtoiarov, I. N., P. M. Sondel, J. C. Eickhoff and A. L. Rakhmilevich (2007).
"Macrophages are essential for antitumour effects against weakly immunogenic
murine tumours induced by class B CpG-oligodeoxynucleotides." Immunology
120(3): 412-423.
Chang, C.-H., P. Gupta and D. M. Goldenberg (2009). "Advances and challenges in
developing cytokine fusion proteins as improved therapeutics."
Chang, Y.-H. (1969). "Studies on phagocytosis: I. Uptake of radio-iodinated (131I) human
serum albumin as a measure of the degree of phagocytosis in vitro." Exp Cell Res
54(1): 42-48.
Chaudhury, C., S. Mehnaz, J. M. Robinson, W. L. Hayton, D. K. Pearl, D. C. Roopenian and
C. L. Anderson (2003). "The major histocompatibility complex-related Fc receptor
for IgG (FcRn) binds albumin and prolongs its lifespan." J Exp Med 197(3): 315-322.
Chauhan, V. P., T. Stylianopoulos, Y. Boucher and R. K. Jain (2011). "Delivery of molecular
and nanoscale medicine to tumors: transport barriers and strategies." Annu Rev Chem
Biomol Eng 2: 281-298.
Cheung, R. and M. Cho (2010). "Importance of avidity for an endogenous drug carrier: an
antibody carrier for CpG oligonucleotides." Mol Pharm 7(4): 1338-1341.
Chung, D. E. and F. Kratz (2006). "Development of a novel albumin-binding prodrug that is
cleaved by urokinase-type-plasminogen activator (uPA)." Bioorg Med Chem Lett
16(19): 5157-5163.
Crooke, S. T. (1992). "therapeutic applications of oligonucleotides."
Crooke, S. T., M. J. Graham, J. E. Zuckerman, D. Brooks, B. S. Conklin, L. L. Cummins, M.
J. Greig, C. J. Guinosso, D. Kornbrust, M. Manoharan, H. M. Sasmor, T. Schleich, K.
L. Tivel and R. H. Griffey (1996). "Pharmacokinetic properties of several novel
oligonucleotide analogs in mice." Journal of Pharmacology and Experimental
Therapeutics 277(2): 923-937.
DeNardo, D. G., J. B. Barreto, P. Andreu, L. Vasquez, D. Tawfik, N. Kolhatkar and L. M.
Coussens (2009). "CD4(+) T cells regulate pulmonary metastasis of mammary
carcinomas by enhancing protumor properties of macrophages." Cancer Cell 16(2):
91-102.
Dunn, G. P., L. J. Old and R. D. Schreiber (2004). "The three Es of cancer immunoediting."
Annu Rev Immunol 22: 329-360.
115
Eder, P. S., R. J. DeVine, J. M. Dagle and J. A. Walder (1991). "Substrate specificity and
kinetics of degradation of antisense oligonucleotides by a 3' exonuclease in plasma."
Antisense Res Dev 1(2): 141-151.
Elsadek, B. and F. Kratz (2012). "Impact of albumin on drug delivery - New applications on
the horizon." J Control Release 157(1): 4-28.
El-Sagheer, A. H., V. V. Cheong and T. Brown (2011). "Rapid chemical ligation of
oligonucleotides by the Diels-Alder reaction." Org Biomol Chem 9(1): 232-235.
Enzler, T., S. Gillessen, J. P. Manis, D. Ferguson, J. Fleming, F. W. Alt, M. Mihm and G.
Dranoff (2003). "Deficiencies of GM-CSF and interferon gamma link inflammation
and cancer." J Exp Med 197(9): 1213-1219.
Folkman, J. (1990). "What is the evidence that tumors are angiogenesis dependent?" J Natl
Cancer Inst 82(1): 4-6.
Funk, W. E., H. Li, A. T. Iavarone, E. R. Williams, J. Riby and S. M. Rappaport (2010).
"Enrichment of cysteinyl adducts of human serum albumin." Anal Biochem 400(1):
61-68.
Graham, M. J., S. M. Freier, R. M. Crooke, D. J. Ecker, R. N. Maslova and E. A. Lesnik
(1993). "Tritium labeling of antisense oligonucleotides by exchange with tritiated
water." Nucleic Acids Res 21(16): 3737-3743.
Graham, M. J., S. T. Crooke, D. K. Monteith, S. R. Cooper, K. M. Lemonidis, K. K. Stecker,
M. J. Martin and R. M. Crooke (1998). "In vivo distribution and metabolism of a
phosphorothioate oligonucleotide within rat liver after intravenous administration."
Journal of Pharmacology and Experimental Therapeutics 286(1): 447-458.
Grulich, A. E., M. T. van Leeuwen, M. O. Falster and C. M. Vajdic (2007). "Incidence of
cancers in people with HIV/AIDS compared with immunosuppressed transplant
recipients: a meta-analysis." The Lancet 370(9581): 59-67.
Gupta, G. K. and D. K. Agrawal (2010). "CpG oligodeoxynucleotides as TLR9 agonists:
therapeutic application in allergy and asthma." BioDrugs 24(4): 225-235.
Hashizume, H., P. Baluk, S. Morikawa, J. W. McLean, G. Thurston, S. Roberge, R. K. Jain
and D. M. McDonald (2000). "Openings between defective endothelial cells explain
tumor vessel leakiness." Am J Pathol 156(4): 1363-1380.
Heckelsmiller, K., K. Rall, S. Beck, A. Schlamp, J. Seiderer, B. Jahrsdorfer, A. Krug, S.
Rothenfusser, S. Endres and G. Hartmann (2002). "Peritumoral CpG DNA elicits a
coordinated response of CD8 T cells and innate effectors to cure established tumors
in a murine colon carcinoma model." J Immunol 169(7): 3892-3899.
Hermanson, G. T. (1996). Bioconjugate techniques. San Diego, Academic Press.
116
Jain, R. K. and T. Stylianopoulos (2010). "Delivering nanomedicine to solid tumors." Nat
Rev Clin Oncol 7(11): 653-664.
Janatova, J., J. K. Fuller and M. J. Hunter (1968). "The heterogeneity of bovine albumin with
respect to sulfhydryl and dimer content." J Biol Chem 243(13): 3612-3622.
Kandimalla, E. R., L. Bhagat, D. Yu, Y. Cong, J. Tang and S. Agrawal (2002). "Conjugation
of ligands at the 5'-end of CpG DNA affects immunostimulatory activity." Bioconjug
Chem 13(5): 966-974.
Kang, Y. S., R. J. Boado and W. M. Pardridge (1995). "Pharmacokinetics and organ
clearance of a 3'-biotinylated, internally [32P]-labeled phosphodiester
oligodeoxynucleotide coupled to a neutral avidin/monoclonal antibody conjugate."
Drug Metab Dispos 23(1): 55-59.
Kawai, T. and S. Akira (2010). "The role of pattern-recognition receptors in innate immunity:
update on Toll-like receptors." Nat Immunol 11(5): 373-384.
Kawarada, Y., R. Ganss, N. Garbi, T. Sacher, B. Arnold and G. J. Hammerling (2001). "NK-
and CD8(+) T cell-mediated eradication of established tumors by peritumoral
injection of CpG-containing oligodeoxynucleotides." J Immunol 167(9): 5247-5253.
Kratz, F. (2007). "DOXO-EMCH (INNO-206): the first albumin-binding prodrug of
doxorubicin to enter clinical trials." Expert Opin Investig Drugs 16(6): 855-866.
Kratz, F., A. Warnecke, K. Scheuermann, C. Stockmar, J. Schwab, P. Lazar, P. Druckes, N.
Esser, J. Drevs, D. Rognan, C. Bissantz, C. Hinderling, G. Folkers, I. Fichtner and C.
Unger (2002). "Probing the cysteine-34 position of endogenous serum albumin with
thiol-binding doxorubicin derivatives. Improved efficacy of an acid-sensitive
doxorubicin derivative with specific albumin-binding properties compared to that of
the parent compound." J Med Chem 45(25): 5523-5533.
Kratz, F., A. Warnecke, K. Scheuermann, C. Stockmar, J. Schwab, P. Lazar, P. Druckes, N.
Esser, J. Drevs, D. Rognan, C. Bissantz, C. Hinderling, G. Folkers, I. Fichtner and C.
Unger (2002). "Probing the cysteine-34 position of endogenous serum albumin with
thiol-binding doxorubicin derivatives. Improved efficacy of an acid-sensitive
doxorubicin derivative with specific albumin-binding properties compared to that of
the parent compound." J Med Chem 45(25): 5523-5533.
Krieg, A. M. (2003). "CpG motifs: the active ingredient in bacterial extracts?" Nat Med 9(7):
831-835.
Krieg, A. M., A. K. Yi, S. Matson, T. J. Waldschmidt, G. A. Bishop, R. Teasdale, G. A.
Koretzky and D. M. Klinman (1995). "CpG motifs in bacterial DNA trigger direct B-
cell activation." Nature 374(6522): 546-549.
117
Krieg, A. M., S. MATSON and E. FISHER (1996). "Oligodeoxynucleotide modifications
determine the magnitude of B cell stimulation by CpG motifs." Antisense and Nucleic
Acid Drug Development 6(2): 133-139.
Krieg, A. M., T. Wu, R. Weeratna, S. M. Efler, L. Love-Homan, L. Yang, A. K. Yi, D. Short
and H. L. Davis (1998). "Sequence motifs in adenoviral DNA block immune
activation by stimulatory CpG motifs." Proc Natl Acad Sci U S A 95(21): 12631-
12636.
Kunikata, N., K. Sano, M. Honda, K. Ishii, J. Matsunaga, R. Okuyama, K. Takahashi, H.
Watanabe, G. Tamura, H. Tagami and T. Terui (2004). "Peritumoral CpG
oligodeoxynucleotide treatment inhibits tumor growth and metastasis of B16F10
melanoma cells." J Invest Dermatol 123(2): 395-402.
Latz, E., A. Schoenemeyer, A. Visintin, K. A. Fitzgerald, B. G. Monks, C. F. Knetter, E.
Lien, N. J. Nilsen, T. Espevik and D. T. Golenbock (2004). "TLR9 signals after
translocating from the ER to CpG DNA in the lysosome." Nat Immunol 5(2): 190-
198.
Latz, E., A. Verma, A. Visintin, M. Gong, C. M. Sirois, D. C. Klein, B. G. Monks, C. J.
McKnight, M. S. Lamphier, W. P. Duprex, T. Espevik and D. T. Golenbock (2007).
"Ligand-induced conformational changes allosterically activate Toll-like receptor 9."
Nat Immunol 8(7): 772-779.
Lau, S., B. Graham, N. Cao, B. J. Boyd, C. W. Pouton and P. J. White (2012). "Enhanced
extravasation, stability and in vivo cardiac gene silencing via in situ siRNA-albumin
conjugation." Mol Pharm 9(1): 71-80.
Leger, R., K. Thibaudeau, M. Robitaille, O. Quraishi, P. van Wyk, N. Bousquet-Gagnon, J.
Carette, J. P. Castaigne and D. P. Bridon (2004). "Identification of CJC-1131-albumin
bioconjugate as a stable and bioactive GLP-1(7-36) analog." Bioorg Med Chem Lett
14(17): 4395-4398.
Liao, W. S., L. S. Jefferson and J. M. Taylor (1986). "Changes in plasma albumin
concentration, synthesis rate, and mRNA level during acute inflammation." Am J
Physiol 251(6 Pt 1): C928-934.
Lin, E. Y. and J. W. Pollard (2004). "Role of infiltrated leucocytes in tumour growth and
spread." Br J Cancer 90(11): 2053-2058.
Loening, A. M. and S. S. Gambhir (2003). "AMIDE: a free software tool for multimodality
medical image analysis." Mol Imaging 2(3): 131-137.
Ma, X. J., S. Dahiya, E. Richardson, M. Erlander and D. C. Sgroi (2009). "Gene expression
profiling of the tumor microenvironment during breast cancer progression." Breast
Cancer Res 11(1): R7.
118
Martin-Armas, M., J. Simon-Santamaria, I. Pettersen, U. Moens, B. Smedsrod and B.
Sveinbjornsson (2006). "Toll-like receptor 9 (TLR9) is present in murine liver
sinusoidal endothelial cells (LSECs) and mediates the effect of CpG-
oligonucleotides." J Hepatol 44(5): 939-946.
Matsumura, Y. and H. Maeda (1986). "A new concept for macromolecular therapeutics in
cancer chemotherapy: mechanism of tumoritropic accumulation of proteins and the
antitumor agent smancs." Cancer Res 46(12 Pt 1): 6387-6392.
Maurer, T., A. Heit, H. Hochrein, F. Ampenberger, M. O'Keeffe, S. Bauer, G. B. Lipford, R.
M. Vabulas and H. Wagner (2002). "CpG-DNA aided cross-presentation of soluble
antigens by dendritic cells." Eur J Immunol 32(8): 2356-2364.
Mroz, P., A. P. Castano and M. R. Hamblin (2009). "Stimulation of dendritic cells enhances
immune response after photodynamic therapy." 717803-717803.
Nierkens, S., M. H. den Brok, T. Roelofsen, J. A. Wagenaars, C. G. Figdor, T. J. Ruers and
G. J. Adema (2009). "Route of administration of the TLR9 agonist CpG critically
determines the efficacy of cancer immunotherapy in mice." PLoS One 4(12): e8368.
Norbury, C. J. and I. D. Hickson (2001). "Cellular responses to DNA damage." Annu Rev
Pharmacol Toxicol 41: 367-401.
Palma, E. and M. J. Cho (2007). "Improved systemic pharmacokinetics, biodistribution, and
antitumor activity of CpG oligodeoxynucleotides complexed to endogenous
antibodies in vivo." J Control Release 120(1-2): 95-103.
Park, B., M. M. Brinkmann, E. Spooner, C. C. Lee, Y. M. Kim and H. L. Ploegh (2008).
"Proteolytic cleavage in an endolysosomal compartment is required for activation of
Toll-like receptor 9." Nat Immunol 9(12): 1407-1414.
Pentlow, K. S., M. C. Graham, R. M. Lambrecht, F. Daghighian, S. L. Bacharach, B.
Bendriem, R. D. Finn, K. Jordan, H. Kalaigian, J. S. Karp, W. R. Robeson and S. M.
Larson (1996). "Quantitative imaging of iodine-124 with PET." J Nucl Med 37(9):
1557-1562.
Peters, T. (1996). All about albumin : biochemistry, genetics, and medical applications. San
Diego, Academic Press.
Piatyszek, M. A., A. Jarmolowski and J. Augustyniak (1988). "Iodo-Gen-mediated
radioiodination of nucleic acids." Anal Biochem 172(2): 356-359.
Pulaski, B. A. and S. Ostrand-Rosenberg (1998). "Reduction of established spontaneous
mammary carcinoma metastases following immunotherapy with major
histocompatibility complex class II and B7.1 cell-based tumor vaccines." Cancer Res
58(7): 1486-1493.
119
Rabkin, C. S., R. J. Biggar and J. W. Horm (1991). "Increasing incidence of cancers
associated with the human immunodeficiency virus epidemic." Int J Cancer 47(5):
692-696.
Rankin, R., R. Pontarollo, X. Ioannou, A. M. Krieg, R. Hecker, L. A. Babiuk, S. van Drunen
and L. V. van den Hurk (2001). "CpG motif identification for veterinary and
laboratory species demonstrates that sequence recognition is highly conserved."
Antisense Nucleic Acid Drug Dev 11(5): 333-340.
Rippe, B., B. I. Rosengren, O. Carlsson and D. Venturoli (2002). "Transendothelial transport:
the vesicle controversy." J Vasc Res 39(5): 375-390.
Rygh, C. B., S. Qin, J. W. Seo, L. M. Mahakian, H. Zhang, R. Adamson, J. Q. Chen, A. D.
Borowsky, R. D. Cardiff, R. K. Reed, F. R. Curry and K. W. Ferrara (2011).
"Longitudinal investigation of permeability and distribution of macromolecules in
mouse malignant transformation using PET." Clin Cancer Res 17(3): 550-559.
Sanchez, A., E. Pedroso and A. Grandas (2012). "Conjugation reactions involving
maleimides and phosphorothioate oligonucleotides." Bioconjug Chem 23(2): 300-
307.
Sands, H., L. J. Gorey-Feret, A. J. Cocuzza, F. W. Hobbs, D. Chidester and G. L. Trainor
(1994). "Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I.
Comparison of a phosphodiester and a phosphorothioate." Mol Pharmacol 45(5): 932-
943.
Schnitzer, J. E. and J. Bravo (1993). "High affinity binding, endocytosis, and degradation of
conformationally modified albumins. Potential role of gp30 and gp18 as novel
scavenger receptors." J Biol Chem 268(10): 7562-7570.
Schnitzer, J. E. and P. Oh (1994). "Albondin-mediated capillary permeability to albumin.
Differential role of receptors in endothelial transcytosis and endocytosis of native and
modified albumins." J Biol Chem 269(8): 6072-6082.
Schnitzer, J. E., A. Sung, R. Horvat and J. Bravo (1992). "Preferential interaction of albumin-
binding proteins, gp30 and gp18, with conformationally modified albumins. Presence
in many cells and tissues with a possible role in catabolism." J Biol Chem 267(34):
24544-24553.
Schoonen, W. G., J. A. De Roos, W. Westerink and E. Débiton (2005). "Cytotoxic effects of
110 reference compounds on HepG2 cells and for 60 compounds on HeLa, ECC-1
and CHO cells.: II Mechanistic assays on NAD (P) H, ATP and DNA contents."
Toxicology in vitro 19(4): 491-503.
Shankaran, V., H. Ikeda, A. T. Bruce, J. M. White, P. E. Swanson, L. J. Old and R. D.
Schreiber (2001). "IFNgamma and lymphocytes prevent primary tumour development
and shape tumour immunogenicity." Nature 410(6832): 1107-1111.
120
Sharma, S., A. L. Dominguez, S. Z. Manrique, F. Cavallo, S. Sakaguchi and J. Lustgarten
(2008). "Systemic targeting of CpG-ODN to the tumor microenvironment with anti-
neu-CpG hybrid molecule and T regulatory cell depletion induces memory responses
in BALB-neuT tolerant mice." Cancer Res 68(18): 7530-7540.
Sharma, S., C. P. Karakousis, H. Takita, K. Shin and S. P. Brooks (2004). "Cytokines and
chemokines are expressed at different levels in small and large murine colon-26
tumors following intratumoral injections of CpG ODN." Neoplasia 6(5): 523-528.
Shirota, H. and D. M. Klinman (2011). "CpG-conjugated apoptotic tumor cells elicit potent
tumor-specific immunity." Cancer Immunol Immunother 60(5): 659-669.
Siegel, R., D. Naishadham and A. Jemal (2013). "Cancer statistics, 2013." CA Cancer J Clin
63(1): 11-30.
Singh, K. V., J. Kaur, G. C. Varshney, M. Raje and C. R. Suri (2004). "Synthesis and
characterization of hapten-protein conjugates for antibody production against small
molecules." Bioconjug Chem 15(1): 168-173.
Society, A. C. (2013). "Cancer Facts & Figures 2013." Atlanta: American Cancer Society.
Srinivasan, S. K., H. K. Tewary and P. L. Iversen (1995). "Characterization of binding sites,
extent of binding, and drug interactions of oligonucleotides with albumin." Antisense
Res Dev 5(2): 131-139.
Stehle, G., H. Sinn, A. Wunder, H. H. Schrenk, J. C. M. Stewart, G. Hartung, W. MaierBorst
and D. L. Heene (1997). "Plasma protein (albumin) catabolism by the tumor itself -
implications for tumor metabolism and the genesis of cachexia." Critical Reviews in
Oncology/Hematology 26(2): 77-100.
Stehle, G., H. Sinn, A. Wunder, H. H. Schrenk, S. Schutt, W. Maier-Borst and D. L. Heene
(1997). "The loading rate determines tumor targeting properties of methotrexate-
albumin conjugates in rats." Anticancer Drugs 8(7): 677-685.
Street, S. E., Y. Hayakawa, Y. Zhan, A. M. Lew, D. MacGregor, A. M. Jamieson, A.
Diefenbach, H. Yagita, D. I. Godfrey and M. J. Smyth (2004). "Innate immune
surveillance of spontaneous B cell lymphomas by natural killer cells and gammadelta
T cells." J Exp Med 199(6): 879-884.
Thibaudeau, K., R. Leger, X. Huang, M. Robitaille, O. Quraishi, C. Soucy, N. Bousquet-
Gagnon, P. van Wyk, V. Paradis, J. P. Castaigne and D. Bridon (2005). "Synthesis
and evaluation of insulin-human serum albumin conjugates." Bioconjug Chem 16(4):
1000-1008.
Tlsty, T. D. and L. M. Coussens (2006). "Tumor stroma and regulation of cancer
development." Annu Rev Pathol 1: 119-150.
121
Vaidyanathan, G. and M. R. Zalutsky (1990). "Protein radiohalogenation: observations on
the design of N-succinimidyl ester acylation agents." Bioconjug Chem 1(4): 269-273.
Verel, I., G. W. Visser, M. J. Vosjan, R. Finn, R. Boellaard and G. A. van Dongen (2004).
"High-quality 124I-labelled monoclonal antibodies for use as PET scouting agents
prior to 131I-radioimmunotherapy." Eur J Nucl Med Mol Imaging 31(12): 1645-
1652.
Williams, D. L., T. Ha, C. Li, J. H. Kalbfleisch, J. J. Laffan and D. A. Ferguson (1999).
"Inhibiting early activation of tissue nuclear factor-κB and nuclear factor interleukin 6
with (1→ 3)-β-d-glucan increases long-term survival in polymicrobial sepsis."
Surgery 126(1): 54-65.
Wolf, A. M., D. Wolf, M. Steurer, G. Gastl, E. Gunsilius and B. Grubeck-Loebenstein
(2003). "Increase of regulatory T cells in the peripheral blood of cancer patients."
Clin Cancer Res 9(2): 606-612.
Xie, D., C. Yao, L. Wang, W. Min, J. Xu, J. Xiao, M. Huang, B. Chen, B. Liu, X. Li and H.
Jiang (2010). "An albumin-conjugated peptide exhibits potent anti-HIV activity and
long in vivo half-life." Antimicrob Agents Chemother 54(1): 191-196.
Xie, X., J. Liang, T. Pu, F. Xu, F. Yao, Y. Yang, Y. L. Zhao, D. You, X. Zhou, Z. Deng and
Z. Wang (2012). "Phosphorothioate DNA as an antioxidant in bacteria." Nucleic
Acids Res 40(18): 9115-9124.
Yamasaki, Y., J. Hisazumi, K. Yamaoka and Y. Takakura (2003). "Efficient scavenger
receptor-mediated hepatic targeting of proteins by introduction of negative charges on
the proteins by aconitylation: the influence of charge density and size of the proteins
molecules." European Journal of Pharmaceutical Sciences 18(5): 305-312.
Yuan, F., M. Dellian, D. Fukumura, M. Leunig, D. A. Berk, V. P. Torchilin and R. K. Jain
(1995). "Vascular permeability in a human tumor xenograft: molecular size
dependence and cutoff size." Cancer Res 55(17): 3752-3756.
Zhang, Q., K. Itagaki and C. J. Hauser (2010). "Mitochondrial DNA is released by shock and
activates neutrophils via p38 map kinase." Shock 34(1): 55-59.